MITOCHONDRIAL DNA RECOVERY AND ANALYSIS FROM
SPENT CARTRIDGE CASINGS
By
Michelle Metchikian

A THESIS
Submitted to
Michigan State University
in partial fulfillment of the requirements
for the degree of
Forensic Science – Master of Science

2013

ABSTRACT
MITOCHONDRIAL DNA RECOVERY AND ANALYSIS FROM
SPENT CARTRIDGE CASINGS
By
Michelle Metchikian
Firearms, particularly handguns, are frequently used in violent crimes, and thus, spent
cartridge casings may be recovered from shooting scenes. Casings have the potential to harbor
DNA deposited by the loader of a weapon, which can be used to produce genetic profiles. While
short tandem repeat (STR) analysis is the standard method of DNA identification in forensic
science today, many researchers have reported limited success obtaining these data from spent
casings. The primary goal of this study was to examine if it is possible to generate mitochondrial
DNA (mtDNA) profiles from spent cartridge casings. Volunteers loaded cartridges into the
magazine of a handgun, the cartridges were fired, and casings were collected and swabbed.
DNAs were extracted, mtDNAs were amplified and sequenced, and haplotypes were generated.
Two swabbing methods, individual and cumulative, were compared to determine which was
most likely to result in mtDNA haplotypes consistent with the loader. Cumulatively swabbing
resulted in a greater frequency of consistent haplotypes, as well as a lower mixture frequency
than individually swabbing. Furthermore, a larger amount of the first loaded/last fired cartridge
casings compared to the first fired cartridges exhibited contamination. Consensus haplotypes,
which contained shared polymorphisms from multiple swabs, yielded a greater percentage of
consistent haplotypes than the individually swabbed spent casings, but did not outperform
cumulatively swabbing. However, none of these differences were statistically significant.
Ultimately, mtDNA analysis is a reliable method to generate genetic profiles recovered from
spent cartridge casings.

ACKNOWLEDGEMENTS

First I would like to thank Dr. David Foran for his assistance throughout this project, his
countless hours of editing, and all the opportunities he has given me. I would also like to thank
Dr. Bryan Epperson and Dr. Carole Gibbs for serving on my committee and taking the time to
learn about my project. A major thanks is in order for Lisa Hebda, Ashley Mottar, Mac Hopkins,
Amanda Fazi, Sarah Rambadt, Drew Fisher, Ashley Doran, Bea List, and Lara Walotka for all of
their assistance in and out of the laboratory. I am very grateful for the monetary support awarded
by the NIJ/FSF Forensic Science Student Research Grant, the Michigan State University (MSU)
Graduate School, and the MSU Forensic Science Program. I thank Alicya Orlando for her DNA
extracts. I also thank Sgt. Ryan Larrison for offering advice about “shooting parties.” I am very
appreciative of Trps. Jon Wickwire and Keyonn Whitfield from the Michigan State Police (MSP)
Sterling Heights Forensic Science Laboratory and Sgts. Dean Molnar and Robert Rayer from the
MSP Northville Forensic Science Laboratory who dedicated their time at work to fire the guns
and help with the collection of casings. I would also like to thank Dr. Jurgen Switalski for
allowing me to collect samples at the Northville Laboratory, for helping me to recruit helpers and
volunteers, and to him and all of the other volunteers who loaded cartridges. Thank you to the
forensic biologists at the Northville Laboratory, which include, but are not limited to Glen Hall,
Brandon Good, David Arnold, Billie Hooker, and Marco Scarpetta for answering questions I had
about their experiences and procedures. I would like to thank Dr. Ryan Kimbirauskas and my
fellow ISB 208L TAs who provided suggestions based on their own research experiences. I
cannot forget to thank Dr. Renate Snider and my fellow classmates from NSC 840 for their edits
and input. Finally, I am very grateful for my family and friends who supported and encouraged

iii

me to keep on kicking through the long Michigan winters, the mosquito infested humid summers,
and particularly through the stresses encountered from classes and research.

iv

TABLE OF CONTENTS

LIST OF TABLES ....................................................................................................................... vii
LIST OF FIGURES ...................................................................................................................... xi
INTRODUCTION ...........................................................................................................................1
Composition of a Cartridge and Ejection of Cartridge Casings ..............................................2
Fingerprints Recovered from Spent Cartridge Casings ...........................................................6
Obtaining Nuclear DNA from Touch Samples .......................................................................7
Utility of nDNA Recovered from Spent Cartridge Casings for Identification of the Loader 10
Processing of Spent Cartridge Casings by Forensic Science Laboratories ...........................12
Mitochondrial DNA from Touch Samples ............................................................................14
Goals of This Study ...............................................................................................................18
MATERIALS AND METHODS ..................................................................................................21
Part A ........................................................................................................................................21
Sequencing Genetic Profiles ..................................................................................................22
Analysis of Genetic Profiles, Consistency of Assignments, and Mixtures ...........................24
Part B ........................................................................................................................................25
Obtaining and Decontaminating Cartridge Casings ..............................................................26
Loading of Cartridges ............................................................................................................26
DNA Recovery and Isolation ................................................................................................27
mtDNA Amplification, Sequencing, Analysis of Genetic Profiles, Consistency of
Assignments, and Mixtures ...................................................................................................29
Construction of Consensus Haplotypes .................................................................................30
Statistical Analyses ................................................................................................................30
RESULTS .....................................................................................................................................32
Part A ........................................................................................................................................32
Degradation of DNA Recovered from Spent Cartridge Casings............................................32
Contamination of Zymo-Spin™ IV-HRC Columns and Cartridges ....................................34
mtDNA Amplification, Sequencing, and Assignments of Orlando (2012) DNAs ...............34
Part B ........................................................................................................................................38
Degradation of DNA Recovered from Spent Cartridge Casings ...........................................38
Contamination of Cartridges and the Magazine ....................................................................39
mtDNA Haplotypes and Loader Assignation for Collections 1 and 2 ..................................40
Detection of Mixtures from Collections 1 and 2 ...................................................................45
Assignments for Collections 1 and 2 .....................................................................................50
DISCUSSION ...............................................................................................................................52
CONCLUSION .............................................................................................................................63

v

APPENDICES ..............................................................................................................................65
Appendix A: mtDNA Profiles from Orlando (2012) ............................................................66
Appendix B: mtDNA Profiles from Collections 1 and 2 ......................................................69
REFERENCES .............................................................................................................................83

vi

LIST OF TABLES

Table 1: Major findings from previous fingerprint and DNA analysis research from spent
cartridge casings. .............................................................................................................................5
Table 2: Primers used to amplify mtDNA from cartridge casings and reference samples. ..........21
Table 3: Nomenclature used to make base calls for all sequences. ..............................................24
Table 4: Frequency of consistent alleles amplified at each locus illustrating DNA degradation of
DNAs from Orlando (2012) extracts that underwent mtDNA analysis. Amplicons are arranged
from smallest to largest for each dye. The smaller amplicons, on average, had higher frequencies
of amplification compared to the larger amplicons. .....................................................................32
Table 5: Pearson’s chi-square p-values comparing assignments for Part A individually and
cumulatively swabbed spent cartridge casings after they were combined. The assignment with
the higher frequency is listed first. There were significantly more consistent-single than
inconsistent, and inconclusive than inconsistent assignments (p = 0.008 and 3.93E-4,
respectively). .................................................................................................................................36
Table 6: The percentage of STR alleles consistent with the loader, number of alleles inconsistent
with the loader, and mtDNA assignments for Orlando (2012) DNA extracts. Extract numbers
followed by an A indicate cumulatively and letters B – F indicate individually swabbed casings.
Extracts with a consistent-single mtDNA assignment had a higher average frequency of
consistent alleles, compared to inconsistent assignments (17.3% vs. 15.3%, respectively). ........37
Table 7: Pearson’s chi-square p-values obtained when comparing assignments made for
combined individually and cumulatively swabbed spent cartridge casings from Collections 1 and
2. The assignment with the higher frequency is listed first. There were significantly more
consistent-combined than inconsistent and inconclusive assignments (6.74E-12 and 1.68E-4,
respectively). There were more inconclusive than inconsistent assignments (3.28E-4). ............45
Table 8: mtDNA assignments made for haplotypes developed with and without MBI for
Collection 1. The fourth column indicates if the extracts contained a mixture that included a
haplotype consistent with the loader (loader mixture). Extract numbers followed by an A
indicate cumulatively swabbed and letters B – D represent individually swabbed spent casings.
MBI did not alter any inconsistent or inconclusive assignments. Six of the 16 extracts contained
a loader mixture. ...........................................................................................................................46
Table 9: mtDNA assignments made for haplotypes developed with and without MBI for
Collection 2. The fourth column indicates if the extracts contained a loader mixture. Extract
numbers followed by an A represent cumulatively swabbed spent casings, and individually
swabbed spent casings are designated by B (first), C (random middle), and D (sixth). MBI

vii

altered 17 of the 35 inconsistent and inconclusive assignments. Fifty-three of the 95 extracts
contained a loader mixture. ...........................................................................................................47
Table 10: Percentage of consistent-combined assignments, inconsistent and inconclusive
assignments, mixture, and loader mixture made for the four Collection 2 categories. The second
and fourth columns summarize the percentage of consistent-combined, and inconsistent and
inconclusive assignemnts made, respectively. The third and fifth columns indicate the
percentage of mixture and loader mixture, respectively, for the two combined assignment types.
The sixth column contains the percentage of the four categories with a loader mixture. The sixth
spent casings had the highest frequency of loader mixture (73.9%), while cumulatively swabbed
casings had the lowest (45.8%). ....................................................................................................50
Table 11: Percentage of consistent-combined, inconsistent, and inconclusive assignments made
from individually, cumulatively, and all swabbed spent casings, as well as All-C and Indiv-C
from Part B. Cumulatively swabbed casings had the highest frequency of consistent-combined
assignments (71.4%), while individually swabbed casings had the lowest frequency (60.2%).
Both Indiv-C and All-C resulted in 67.9% consistent-combined assignments. ............................51
Table A1: mtDNA profiles obtained from spent cartridge casings loaded by volunteer H. .........65
Table A2: mtDNA profiles obtained from spent cartridge casings loaded by volunteer K. .........65
Table A3: mtDNA profiles obtained from spent cartridge casings loaded by volunteer O. .........66
Table A4: mtDNA profiles obtained from spent cartridge casings loaded by volunteer P. .........66
Table A5: mtDNA profiles obtained from spent cartridge casings loaded by volunteer R. .........66
Table A6: mtDNA profiles obtained from spent cartridge casings loaded by volunteer T. .........67
Table A7: mtDNA profiles obtained from spent cartridge casings loaded by volunteer U. .........67
Table A8: mtDNA profiles obtained from spent cartridge casings loaded by volunteer DD. ......67
Table B1: mtDNA profiles obtained from spent cartridge casings loaded by volunteer M during
Collection 1. ..................................................................................................................................68
Table B2: mtDNA profiles obtained from spent cartridge casings loaded by volunteer S during
Collection 1. ..................................................................................................................................69
Table B3: mtDNA profiles obtained from spent cartridge casings loaded by volunteer HH during
Collection 1. ..................................................................................................................................69
Table B4: mtDNA profiles obtained from spent cartridge casings loaded by volunteer OO during
Collection 1. ..................................................................................................................................70

viii

Table B5: mtDNA profiles obtained from spent cartridge casings loaded by volunteer B during
Collection 2. ..................................................................................................................................70
Table B6: mtDNA profiles obtained from spent cartridge casings loaded by volunteer C during
Collection 2. ..................................................................................................................................71
Table B7: mtDNA profiles obtained from spent cartridge casings loaded by volunteer E during
Collection 2. ..................................................................................................................................71
Table B8: mtDNA profiles obtained from spent cartridge casings loaded by volunteer H during
Collection 2. ..................................................................................................................................72
Table B9: mtDNA profiles obtained from spent cartridge casings loaded by volunteer I during
Collection 2. ..................................................................................................................................72
Table B10: mtDNA profiles obtained from spent cartridge casings loaded by volunteer J during
Collection 2. ..................................................................................................................................73
Table B11: mtDNA profiles obtained from spent cartridge casings loaded by volunteer L during
Collection 2. ..................................................................................................................................73
Table B12: mtDNA profiles obtained from spent cartridge casings loaded by volunteer N during
Collection 2. ..................................................................................................................................74
Table B13: mtDNA profiles obtained from spent cartridge casings loaded by volunteer T during
Collection 2. ..................................................................................................................................74
Table B14: mtDNA profiles obtained from spent cartridge casings loaded by volunteer BB
during Collection 2. ......................................................................................................................75
Table B15: mtDNA profiles obtained from spent cartridge casings loaded by volunteer EE
during Collection 2. ......................................................................................................................75
Table B16: mtDNA profiles obtained from spent cartridge casings loaded by volunteer FF during
Collection 2. ..................................................................................................................................76
Table B17: mtDNA profiles obtained from spent cartridge casings loaded by volunteer GG
during Collection 2. ......................................................................................................................76
Table B18: mtDNA profiles obtained from spent cartridge casings loaded by volunteer II during
Collection 2. ..................................................................................................................................76
Table B19: mtDNA profiles obtained from spent cartridge casings loaded by volunteer JJ during
Collection 2. ..................................................................................................................................77

ix

Table B20: mtDNA profiles obtained from spent cartridge casings loaded by volunteer NN
during Collection 2. ......................................................................................................................77
Table B21: mtDNA profiles obtained from spent cartridge casings loaded by volunteer PP during
Collection 2. ..................................................................................................................................78
Table B22: mtDNA profiles obtained from spent cartridge casings loaded by volunteer RR
during Collection 2. ......................................................................................................................78
Table B23: mtDNA profiles obtained from spent cartridge casings loaded by volunteer SS during
Collection 2. ..................................................................................................................................79
Table B24: mtDNA profiles obtained from spent cartridge casings loaded by volunteer TT
during Collection 2. ......................................................................................................................79
Table B25: mtDNA profiles obtained from spent cartridge casings loaded by volunteer UU
during Collection 2. ......................................................................................................................80
Table B26: mtDNA profiles obtained from spent cartridge casings loaded by volunteer VV
during Collection 2. ......................................................................................................................80
Table B27: mtDNA profiles obtained from spent cartridge casings loaded by volunteer WW
during Collection 2. ......................................................................................................................81
Table B28: mtDNA profiles obtained from spent cartridge casings loaded by volunteer XX
during Collection 2. ......................................................................................................................81

x

LIST OF FIGURES

Figure 1: Anatomy of a rimfire and centerfire cartridge. Adopted from
http://www.gunclassics.com/ammunition.html. .............................................................................3
Figure 2: Flow chart of the collection of samples and isolation of DNA used in the Orlando
(2012) study. ...................................................................................................................................9
Figure 3: The human mtDNA genome. Taken from National Forensic Science Technology
Center. Available at http://www.nfstc.org/pdi/Subject09/pdi_s09_m01_01_b.htm. ...................15
Figure 4: Sequencing electropherograms from buccal swabs of two individuals (top and middle)
and a mixture of the mtDNA sequences from the two individuals (bottom). The positions
showing mixture are indicated with a ‘Y’ (pyrimidine) above the peaks. ....................................17
Figure 5: Electropherogram containing a heteroplasmic position indicated with a ‘Y’ above the
peaks. The buccal sample of this individual possessed a T (red) peak and a C (blue) peak. ......18
Figure 6: a. Cartridge casings individually double-swabbed on the left and cumulatively doubleswabbed on the right. b. Flow chart of the collection of casings and recovery of cells from
Collection 1. c. Flow chart of the collection of casings and recovery of cells from Collection 2. ..
........................................................................................................................................................28
Figure 7: Four percent agarose gel electrophoreses showing PCR products of extracts 22D and
28C from Orlando (2012). Primer pairs were: a. HV1 (left), HV1b (center), HV2a (right), and b.
HV2. No bands were present in the negative controls (Neg), and were in the positive controls
(Pos). The extracts amplified using HV1b and HV2a, had decreased amplification using HV1,
and no amplification using HV2. ..................................................................................................33
Figure 8: Four percent agarose gel electrophoresis showing PCR results from the Zymo-Spin™
IV-HRC column contamination test. No bands were present in the negative controls and filtrate
T2 using HV1a and HV2a, but there were bands for T1. .............................................................34
Figure 9: Assignments made for Orlando (2012) DNA extracts: a. individually swabbed and b.
cumulatively swabbed spent cartridge casings. The majority of assignments for both swab
methods were inconclusive. ..........................................................................................................35
Figure 10: a. Four percent agarose gel electrophoreses showing PCR results of Collection 1
extract 12C using primer pairs HV1a, HV2a, b. HV1, and HV2. In 8a., PCR results for extracts
12A – D are shown; 12C extracts are boxed in red. No bands were present in negative controls
and the reagent blank (RB). Extract 12C amplified using HV1a and HV2a, had a lower level of
amplification using HV2, and no amplification using HV1. ........................................................39

xi

Figure 11: Four percent gel electrophoresis showing PCR results from live cartridges L1 – 3
using primer pairs HV1a and HV2a. L1 – 3 amplified, while no bands were present in the
reagent blanks or negative controls. ..............................................................................................40
Figure 12: Assignments made for Collection 1: a. individually swabbed and b. cumulatively
swabbed spent cartridge casings. The majority of assignments for both swab methods were
consistent-single. ...........................................................................................................................41
Figure 13: Assignments made for spent casings from Collection 2: a. the first (n = 24), b. a
random middle (n = 24), c. the sixth (n = 23), and d. three cumulatively swabbed spent cartridge
casings (n = 24). When consistent-single and consistent-multiple assignments were combined,
they comprised 50.0% or more of assignments for the four categories. .......................................43
Figure 14: Assignments made for all profiles from Collections 1 and 2; consistent-single and
consistent-multiple assignments were combined. The majority of assignments for all Part B
extracts were consistent-combined (63.1%). ................................................................................44

xii

INTRODUCTION

The Second Amendment to the Constitution of the United States, ratified in 1791, grants
citizens the right to own firearms. Almost a century and a half elapsed before startling events
involving firearms (e.g., the St. Valentine’s Day Massacre in 1929 and the assassination of
President John F. Kennedy in 1963) became catalysts for new gun control laws (Bingham, 2012;
Goforth, 2013). Despite the implementation of some stringent laws, the 310 million guns
possessed both legally and illegally in the US (Krouse, 2012) are not going anywhere. This is
because “they last several human lifetimes with minimal maintenance” (Dulan, 2012). Since
there are so many firearms, it should be no surprise that they are commonly used weapons in
crimes. According to the 2012 Federal Bureau of Investigation (FBI) Uniform Crime Report,
67.8% of the 14,612 murders, 41.3% of the 145,366 robberies, and 21.2% of the 159,240
aggravated assaults in the US in 2011 involved firearms. They were also used as an accessory in
forcible rapes (although data on weapons are not collected after such incidences). Furthermore,
19,766 suicides were committed with a firearm in 2011 (Hoyert and Xu, 2012).
Since the early 1990s, crime in the US has steadily declined, with a 17% decrease in the
murder rate from 2001 – 2011 (Frieden, 2012). However, the rate of unsolved murders increased
from 7% in 1964 (Richardson and Kosa, 2001) to approximately 30% in 2011 (FBI, 2012). This
rise is largely due to a higher frequency of stranger-to-stranger homicides as opposed to
acquaintance homicides, making it more difficult for investigators to determine a motive and
connection to the victim. While acquaintance homicides typically result from a conflict between
the victim and offender, felonies, particularly robberies, are committed concurrently with a
stranger-to-stranger homicide. Since it is common for perpetrators to plan the robberies, the

1

chance of investigators discovering physical evidence from a stranger-to-stranger homicide is
decreased, as extra care is given by the perpetrators to minimize evidence left behind
(Richardson and Kosa, 2001). Perpetrators may clean up after themselves, for example by
collecting casings ejected from a firearm. Furthermore, 72.5% of all firearm homicides involve
the use of a handgun, despite the fact that they make up only 37% of the firearms in the US (FBI,
2012). This high frequency of handgun usage is attributed to their ease of use and ability to
conceal because they do not possess long barrels like rifles and shotguns.

Composition of a Cartridge and Ejection of Cartridge Casings
A cartridge consists of a bullet, propellant powder, primer, and a casing (Figure 1). The
role of the primer, which is comprised of an explosive initiation compound, fuel, and oxidizer
(Krampen et al., 1986), is to start a fire and deflagrate the nitrocellulose-based propellant powder.
The primer is contained either in the rim or center of the casing head, which ignites when the
firing pin strikes (Westrom et al., 2005). The buildup of gases caused by the burning of the
propellant expands the casing and forces the bullet out of the cartridge and through the barrel of
the firearm. The casing is extracted and ejected from a semi-automatic and automatic firearm by
the action mechanism (Sparano, 2000), resulting in a key piece of physical evidence.

2

Figure 1. Anatomy of a rimfire and centerfire cartridge. Adopted from
http://www.gunclassics.com/ammunition.html. For interpretation of the references to color in
this and all other figures, the reader is referred to the electronic version of this thesis.

Unless the shooter collected the spent cartridge casing(s) or used a firearm that does not
eject casings (e.g. a revolver), casings will be left behind at the scene of a shooting. The marks
made by the firing pin, breechface, extractor, and ejector on the surface of a casing are
considered individualizing characteristics that can be used to determine the firearm a casing was
ejected from (Cork et al., 2010). However, this comparison can only be made if known casings
from the firearm are available (Milloy, 2002). Furthermore, this method cannot be used to
identify the loader of the firearm; individualizing evidence from a spent casing is contingent on
recovering fingerprints and/or ample amounts of DNA left behind. Even if the loader did not
pull the trigger, their identification would help with the investigation. For instance, when the
identified loader is questioned, the individual may disclose who used or had access to the loaded

3

firearm. Unfortunately, these identification methods are not always met with success; forensic
scientists seldom attempt fingerprint and DNA analyses from the casing surfaces because viable
results are rarely obtained (David Arnold and Brandon Good, personal communications).
Researchers have examined what happens to identifiable materials left on cartridge casings, with
the goal of determining a reliable technique to be used for successful identification. A summary
of the major findings from some of the notable studies is outlined in Table 1.

4

Table 1. Major findings from previous fingerprint and DNA analysis research from spent cartridge casings.
Study
Given
(1976)
Bentsen et al.
(1996)
Spear et al.
(2005)
Horsman-Hall et al.
(2009)

Fingerprint Analysis
 Gaseous blowback destroys
ridge detail
 Friction from loading and
ejection destroy ridge detail
 1 of 24 spent casings with a
viable print (bloody)

DNA Quantification


-

Branch
(2010)






Orlando
(2012)
-

-

-

-

-



STR Analysis

Significantly > DNA recovered
using DNA IQ™ vs. organic
extraction
Average 0.42 ng from spent
casings
DNA recovered from breechface,
chamber, and ejection port
(surfaces)








Quantified if ≥ 3 correct alleles

amplified using MiniFiler™ (n =
47; no distinction of firing status)
11 had ≥ 0.1 ng
No difference between

individually (7.67 ± 9.55 pg/μL)
and cumulatively (38.23 ± 141.52
pg/μL) swabbed spent casings


- indicates no research conducted
5

1 of 24 spent casings with a partial
profile (bloody)
Significantly > alleles using
MiniFiler™ vs. PowerPlex® 16
BIO
No amplification using Identifiler®
Loader alleles recovered from
surfaces
Individually swabbed (n = 75)
Full profiles (1%)
Partial profiles (45%)
No profiles (54%)
Individually swabbed (n = 251)
Full profiles (0%)
Partial profiles (8%)
No profiles (92%)
Individually swabbed (n=155)
Full profiles (3.2%)
Partial profiles (89.7%)
No profiles (7.1%)
Cumulatively swabbed (n=31)
Full profiles (0.0%)
Partial profiles (93.5%)
No profiles (6.5%)

Fingerprints Recovered from Spent Cartridge Casings
Whenever two objects come into contact, an exchange is thought to occur (Locard, 1928).
For example, when a person touches the door of a dirty car, fingerprints and cells containing
DNA are transferred to the door and dirt from the car is transferred to the person’s hand.
Therefore, if gloves are not worn, it is probable that fingerprints will be deposited on cartridge
casings when an individual loads a firearm, which could subsequently be used to make an
identification.
Given (1976) investigated the retention of fingerprints on casings after firing; volunteers
picked up 108 brass and 108 nickel-plated .38-caliber cartridges and placed them on a tray. Half
the cartridges were fired, casings were dusted, and prints were lifted. Discoloration on a portion
of the casing indicated gaseous blowback occurred along the side that is not sealed against the
wall of the chamber during firing. As a result of this blowback, fingerprints deteriorated.
Bentsen et al. (1996) also investigated fingerprint recovery from spent casings by rolling
fingerprints on various ammunition and firing with different types of guns (total of 21
combinations). Ridge detail was recovered from 17 casings, only five (23.8%) of which were
identifiable. While they did find that gaseous blowback destroyed ridge detail, they concluded
that friction from loading the cartridges and during ejection of the spent casings were the main
contributors to loss of detail. This conclusion was reached because four of the five identifiable
prints came from cartridges that had been gently loaded and extracted from revolvers in a
manner that eliminated the added friction encountered with manual firearms. Spear et al. (2005)
studied the feasibility of recovering fingerprints from spent casings, by placing eccrine, oily, or
bloody prints on 48 cartridges (.22 – .45-caliber) made with aluminum, brass, or nickel-plated
brass casings. Half of the cartridges were fired and all casings were then stored at room

6

temperature for several months. Oily and eccrine prints were treated by cyanoacrylate fuming
and rhodamine 6G, and bloody prints with amido black. Prints were recovered from five live
cartridges (20.8%). The only print recovered from a spent casing was made with blood, which is
atypical in that an individual’s hands would not normally have blood on them when loading a
gun. Furthermore, the Indianapolis Metropolitan Police Department East District processed 201
cartridges and casings for fingerprints and only one viable print was developed (Nunn, 2013).
Although it is not mentioned whether the print was from a spent or live cartridge casing, it is
indisputable that fingerprints can rarely be obtained from either, both in a controlled laboratory
setting and in casework.

Obtaining Nuclear DNA from Touch Samples
In 1997, van Oorschot and Jones reported that nuclear DNA (nDNA) profiles could be
generated from briefly handled objects (touch samples). Thus, when an individual loads
cartridges either into a magazine or the chamber of a firearm, cells are likely to be deposited on
the casings. However, a limitation of touch samples is that a variable number of cells is shed
onto a surface—often resulting in less than 0.30 nanogram of DNA (Schulz and Reichert, 2002).
In addition, the amount of cells transferred may differ substantially among loaded cartridges.
For instance, most cells might be deposited on the first cartridge loaded. In contrast, because an
increased amount of force is required to push down on the spring as more cartridges are loaded,
the last cartridge might harbor the greatest number of cells. Horsman-Hall et al. (2009)
recovered an average of 0.54 ± 0.17 ng of DNA from five unfired 7.62 mm X 39 mm cartridges
handled by volunteers for 30 s. However, because these rifle cartridges were longer than those

7

designed for handguns, it is possible that more cells were transferred onto the surface than would
be on smaller cartridges.
Based on the finding that ridge detail is lost from blowback during firing (Given, 1976)
and also during loading of a cartridge and ejection of a casing (Bentsen et al., 1996), it can be
rationalized that cells are also lost during these processes. Therefore, it is expected that less
DNA would be recovered from spent cartridge casings compared to live cartridge casings.
Horsman-Hall et al. (2009) did recover a slightly lower average quantity of DNA from five spent
casings (0.42 ± 0.10 ng), but the difference was not significant, likely due to the small sample
size. In a separate part of the same study, the ejection port, breechface, and chamber of a
shotgun were double-swabbed using a wet followed by a dry swab, after which 10 volunteers
loaded shotgun shells into the gun (no DNA was recovered from the surfaces prior to firing).
After the shotgun shells from the third (set 1), sixth (set 2), and tenth (set 3) volunteers were
fired, the surfaces were swabbed again, resulting in an average of 0.07 ± 0.07 ng of DNA
obtained from the ejection port, 0.06 ± 0.10 ng from the breechface, and 0.51 ± 0.14 ng from the
chamber of the shotgun. These data also indicate that DNA is lost from shells during firing.
However, the authors did not investigate whether the order of loading influenced DNA
deposition on shells and casings.
Orlando (2012) investigated whether double-swabbing multiple spent casings
successively (cumulatively), as opposed to individually, increased the amount of DNA recovered.
First, ten cartridges were selected at random out of the manufacturer’s box and tested for nDNA;
no background DNA was quantified from the cartridge casings. Volunteers then loaded 10
cartridges into the magazine of a gun and the cartridges were fired. DNA was recovered
individually from 5 of the 10 spent casings using a double-swab method for each casing. The

8

remaining five spent casings were cumulatively swabbed with one pair of swabs. An organic
extraction was performed followed by DNA concentration and purification with Amicon® and
Zymo-Spin™ columns. A flow diagram of sample collection and DNA isolation used by
Orlando (2012) is shown in Figure 2. DNAs were then quantified using a Quantifiler™ Human
DNA Quantification kit. There was no statistical difference between DNA yields obtained using
the two recovery methods.

Volunteers loaded 10 cartridges

10 spent casings per volunteer
collected together
r

Random 5 double-swabbed
individually

Random 5 double-swabbed
cumulatively

Organic extraction

DNA concentration using Amicon® Ultra-0.5 mL columns

Further purification using Zymo-Spin™ IV-HRC columns
Figure 2. Flow chart of the collection of samples and isolation of DNA used in the Orlando
(2012) study.

Horsman-Hall et al. (2009) also compared extraction methods of DNA recovered from
spent cartridge casings. Unfired cartridges handled for 30 s were double-swabbed and DNAs
were digested and extracted using one of four protocols: (1) proteinase K and 20% sarkosyl

9

digestion and DNA IQ™ extraction; (2) digestion with DNA IQ™ lysis buffer, followed by a
DNA IQ™ extraction; (3) proteinase K and sodium dodecyl sulfate (SDS) digestion with SDS
added to a DNA IQ™ extraction (final SDS concentrations unknown); and (4) proteinase K and
20% sarkosyl digestion and an organic extraction using Microcon® purification. The organic
extraction recovered significantly less DNA than the DNA IQ™ methods, among which there
was no significant difference.

Utility of nDNA Recovered from Spent Cartridge Casings for Identification of the Loader
The genetic material deposited on a casing can potentially be used for identification. An
identification using nDNA in forensic applications is made by examining short tandem repeats
(STRs), which are loci that have a variable number of repeated DNA sequences that are typically
2 to 6 base pairs (bp) long. Individuals have two alleles at every locus: one inherited from each
parent. About 0.5 – 1.0 ng of non-degraded DNA is required to obtain a full STR profile (Gill,
2001). If lower quantities of DNA are present, stochastic sampling effects can occur, wherein
one allele is unequally amplified over the other, giving false homozygous results.
Regardless of the quantity of DNA recovered from spent cartridge casings, it is important
to determine if STR profiles can be generated from them, since full profiles have been obtained
from as little as 0.06 ng of nDNA (approximately 10 cells worth) using AmpFℓSTR®
MiniFiler™ (Viray, 1998). In addition to examining the utility of fingerprint analysis from live
and spent casings, Spear et al. (2005) investigated DNA analysis from the casings using a wet
swab, extracted DNA organically, and amplified STRs using Profiler™ Plus. Only one spent

10

casing resulted in STR amplification (4.2%); the profile contained 9 of the 10 loci targeted by the
kit.
Horsman-Hall et al. (2009) also examined STRs from five types of handled ammunition,
including two handgun cartridges with different casing metals, two rifle cartridges with different
casing metals, and shotgun shells, and compared STR results obtained using AmpFℓSTR®
Identifiler®, MiniFiler™, and PowerPlex® 16 BIO kits. The ammunitions were held for 30 s, a
gloved firearms examiner loaded them into the chamber of a firearm, and fired them. The spent
casings and shells were double-swabbed and since there was no significant difference among the
DNA IQ™ extraction methods (detailed above), the authors used the standard protocol of the
Virginia Department of Forensic Science to digest and purify the DNAs. DNAs were quantitated
using Plexor® HY System, and STRs were amplified. No alleles were obtained using the
Identifiler® kit, and there was a significantly greater number of amplified alleles using the
MiniFiler™ kit, designed for degraded samples, compared to the PowerPlex® 16 BIO kit. Only
1% of the fired casings generated full profiles with the MiniFiler™ kit, while 54% of the
casings had no alleles amplify.
Branch (2010) also investigated the utility of MiniFiler™ and Identifiler® for analyzing
DNA from spent casings, but used a robotic extraction method and examined several variables.
Ten volunteers washed their hands, immediately handled .22-caliber brass cartridges for 15, 30,
or 60 s, and the live cartridges were collected. This protocol was repeated, but cartridges were
loaded into a firearm, fired, and spent casings were collected. In phase 2 of the study, five
volunteers waited one hour after washing their hands before handling .38-caliber and 9 mm brass,
nickel, and aluminum cartridges. The cartridges were handled for the same amount of time and

11

collected in the same manner as in phase 1, resulting in 420 samples for phase 2. The casings
were double-swabbed, a BioRobot®EZ1 robot was used to extract DNA, and STRs were
amplified using MiniFiler™. Extracts with a minimum of three amplified alleles consistent
with the volunteer were quantified using a Quantifiler™ Human DNA Quantification kit and
extracts containing a minimum of 0.1 ng of DNA were amplified using Identifiler®. Caliber,
metal type, and handling duration did not have significant effects on the number of alleles
amplified. Ninety-two percent of fired casings resulted in no amplified alleles and significantly
more alleles were obtained when casings were handled one hour after washing hands. Since the
majority of the DNA recovered from live handgun cartridges yielded no profiles (80%), this
further supports the theory that an insufficient amount of DNA is recovered from casings for
successful STR analysis.
Orlando (2012) used AmpFℓSTR® Identifiler® Plus to generate STR profiles from
DNAs recovered from fired casings and compared the number of alleles obtained between
individually and cumulatively swabbed casings to determine if it was advantageous to
cumulatively swab. The majority of the individually and cumulatively swabbed casings (74.2%
vs. 67.7%) had seven or fewer Identifiler® Plus loci amplify, though the frequencies were not
dependent on swab method. Consistent with previous studies (Spear et al., 2005; Horsman-Hall
et al., 2009; Branch, 2010), Orlando (2012) showed that DNA was recoverable from spent
cartridge casings, but STR success was low.

Processing of Spent Cartridge Casings by Forensic Science Laboratories
When spent cartridge casings are collected at a crime scene by the Miami-Dade Police
Department, each casing is packaged separately in a coin envelope and sent to a latent print

12

examiner at the Forensic Laboratory, where the examiner visually inspects the casings to
determine if any prints can be lifted (Forensic Scientist David Arnold, personal communication).
The casings are then transferred to the biology/DNA unit and a DNA analyst processes them
only if no other DNA evidence is available for the case, because as noted, STR profiles from
spent cartridge casings are rarely obtained. All casings from the same investigation are swabbed
with one swab wetted with deionized water, followed by an organic extraction and DNA
quantification. A minimum of 0.35 ng of DNA is required to proceed with STR amplification
using the Identifiler® kit.
Similarly, Michigan State Police (MSP) routinely package spent casings separately
(Forensic Scientist Brandon Good, personal communication). At the MSP Northville Forensic
Science Laboratory, the firearms unit examines the casings to determine if they were likely
expelled from the same weapon. The latent print and biology units only process the casing(s) if a
request, e.g. by the prosecutor, has been made to the forensic science division command. When
casings are processed, the biology unit individually double-swabs them, unless the firearms unit
determines that they were fired from the same gun, in which case they are cumulatively doubleswabbed. DNA is extracted organically, purified using Vivacon® 2 columns, and DNA is
quantified using Plexor® HY. However, unlike the Miami-Dade Police Department, MSP still
proceed to amplify STRs using PowerPlex® 16 BIO, even if no DNA was detected by
quantification.
Studies on STR profiling from spent cartridge casings in casework are quite limited. An
exception is retrospective research conducted by Dieltjes et al. (2011) at the Forensic Laboratory
for DNA Research in the Netherlands, where 4,085 live cartridges, bullets, and casings collected
from 616 cases were processed. STR profiles with at least one amplified locus were obtained

13

from 283 of the items, with an average amplification rate of 10.95 loci per item using the
PowerPlex® 16 system. Forty-four of these items resulted in a full profile from a single donor.
However, there were two major limitations of the study: (1) the type of evidence from which
STR data were obtained was not specified and (2) the loader’s DNA profile was not known and
therefore, it cannot be determined if the amplified alleles from the 48 items were consistent with
the loader.

Mitochondrial DNA from Touch Samples
Human mitochondrial DNA (mtDNA), consisting of approximately 16,569 bp, has 37
genes: 22 transfer RNA genes, 13 protein-coding genes, and 2 ribosomal RNA genes (Anderson
et al., 1981). The control region, depicted at the top of Figure 3, commonly harbors differences
among unrelated individuals. Two hypervariable regions (HV1 and HV2) within the control
region are sequenced in forensic applications. The sequences are compared to the Cambridge
Reference Sequence (Anderson et al., 1981) to locate polymorphisms. The possessed
polymorphisms comprise an individual’s haplotype (mtDNA profile).

14

Figure 3. The human mtDNA genome. Taken from National Forensic Science Technology
Center. Available at http://www.nfstc.org/pdi/Subject09/pdi_s09_m01_01_b.htm.

There is a greater likelihood of successfully obtaining DNA typing results using mtDNA
compared to nDNA (Goodwin et al., 1999). Three factors contribute to the higher success rate.
First, the high copy numbers of mtDNA in a cell increase the probability of isolating DNA. A

15

single cell may contain hundreds of mitochondria, each with an average of 4.6 mtDNA
molecules (Robin and Wang, 1988; Satoh and Kuroiwa, 1991). Second, the linear structure of
nDNA potentially makes it susceptible to degradation by exonucleases, whereas mtDNA’s
circular structure would make exonuclease attack more difficult. Third, the location of mtDNA
molecules in the mitochondrion has been shown to better protect mtDNA compared to DNA in
the nucleus (Foran, 2006).
Since mtDNA is maternally inherited, individuals exhibit a high degree of
homoplasmy—only one mtDNA type (Monnat and Loeb, 1985; Monnat and Reay, 1986;
Bodenteich et al., 1991). Therefore, mtDNA sequences are easily interpretable, unless either: (1)
a mixture of DNA from multiple individuals exists or (2) heteroplasmy is exhibited. A mixture
may result from secondary transfer. For instance, when two people shake hands, an exchange of
DNA may occur (van Oorschot and Jones, 1997). When one of the individuals then proceeds to
touch a (e.g.) pen, it is possible that genetic information from both of them will be deposited onto
the surface (Goray et al., 2010). A mixture can also result from contamination, which can occur
at any step from collection to processing of the evidence. A mtDNA mixture is identified by
observing a clean sequence with more than one peak at certain positions, attributed to more than
one individual (exemplified in Figure 4).

16

Individual 2 Individual 1
Mixture
Figure 4. Sequencing electropherograms from buccal swabs of two individuals (top and middle)
and a mixture of the mtDNA sequences from the two individuals (bottom). The positions
showing mixture are indicated with a ‘Y’ (pyrimidine) above the peaks.

An estimated 2 – 8% of the human population is heteroplasmic—harboring more than
one mtDNA variant (Holland and Parsons, 1999). Four scenarios can result in heteroplasmy: (1)
different cell types (e.g. blood cells and bone cells) with different mtDNA variants; (2) same cell
type with different mtDNA variants; and any cells that contain two variants, either (3) among
mitochondria or (4) within mitochondria (Wilson et al., 1997). Two peaks at one position of an
electropherogram may be indicative of heteroplasmy, such as in Figure 5. The presence of a T
and C peak is due to two mtDNA variants at the fourth position shown. Contrary to intuition, the
proportion of the bases at a heteroplasmic site depicted in an electropherogram is not always

17

indicative of the actual ratio because peak heights have been observed to differ between the
forward and reverse strands of the same extract (Sigursardottir et al., 2000); the same would hold
true for mixtures containing multiple sources of DNA. Furthermore, the mtDNA ratio can differ
between cells. For example, epithelial cells from a buccal sample may contain a different ratio
than the epithelial cells left behind on a cartridge casing. Moreover, one of the epithelial cell
origins may contain only one mtDNA variant.

Figure 5. Electropherogram containing a heteroplasmic position indicated with a ‘Y’ above the
peaks. The buccal sample of this individual possessed a T (red) peak and a C (blue) peak.

Goals of This Study
The ability to use DNA recovered from a spent cartridge casing(s) is vital to identifying a
criminal, especially in instances when other physical evidence from a crime scene is not
available. Despite the fact that scientists have tested various swabbing strategies, DNA
extraction techniques, and STR kits, STR typing success remains low. To date, assaying mtDNA
from spent cartridge casings has not been presented in the literature or at scientific conferences,
most likely at least in part because the majority of forensic science laboratories do not perform
mtDNA analysis. However, in 2005, the FBI partnered with the Arizona Department of Public

18

Safety Central Crime Laboratory, the Connecticut State Police Forensic Science Laboratory, the
Minnesota Bureau of Criminal Apprehension Forensic Science Laboratory, and the New Jersey
State Police Crime Laboratory to provide a free mtDNA analysis service to local and state law
enforcement agencies nationwide (FBI, 2006). Access to mtDNA processing facilities allows
analysis of DNA recovered from spent cartridge casings to be a feasible option across the
country. Thus, if mtDNA analysis from such samples yields usable information when STR
analysis does not, it is a resource that should be taken advantage of.
The primary goal of the research presented here was to thoroughly examine the feasibility
of obtaining mtDNA profiles from spent cartridge casings that are consistent with the loader.
Results from the first and last loaded cartridges were compared to assess which yields haplotypes
consistent with the loader at a higher frequency, coupled with a lower frequency of mixture.
This assessment was also made for haplotypes recovered from individually and cumulatively
swabbed casings. Finally, a consensus profiling method, combining shared polymorphisms from
multiple swabs, was developed to evaluate if this would filter out polymorphisms inconsistent
with the loader and subsequently yield more consistent haplotypes.
The experiment was organized into two parts: Part A and B. Orlando (2012) DNA
extracts were processed for mtDNA in Part A and new samples were collected and processed for
mtDNA in Part B. The following research questions were addressed:
Q1. Can mtDNA haplotypes consistent with the loader be developed from DNA left behind
on cartridge casings during loading?
Q2. How do mtDNA results from spent casings compare to STR results?
Q3. Does the order cartridges are loaded/fired influence mtDNA results?
Q4. Are there differences in mtDNA results from individually and cumulatively double-

19

swabbed spent casings?
Q5. Is it advantageous to use consensus profiling (generating a haplotype comprised of
polymorphisms seen the majority of the time from casings handled by the same
individual)?

20

MATERIALS AND METHODS

Part A
Eleven mtDNA control region primer pairs, including F15989 – R16410 (HV1), F15 –
R484 (HV2), F15989 – R16322, F15989 – R16251, F16057 – R16322, F155 – R484, F155 –
R389, and F256 – R484, were tested on Orlando (2012) DNA extracts 22D and 28A – F. The
three primer pairs subsequently used for this study were: F15989 – R16233 (HV1a), F16190 –
R16410 (HV1b), and F15 – R285 (HV2a) (Table 2). The entire control region of reference
samples was amplified using F15989 – R484. F15989, F16190, R16410, F15, and R285 were
developed at the Armed Forces DNA Identification Laboratory (Edson et al., 2004) and R16233
was developed by Lee et al. (2008).

Table 2. Primers used to amplify mtDNA from cartridge casings and reference samples.
HV1a
F15989
R16233
5’ CCC AAA GCT AAG ATT CTA AT 3’
5’ TGA TAG TTG AAG GTT GAT TGC TGT 3’
HV1b
F16190
R16410
5’ CCC CAT GCT TAC AAG CAA GT 3’
5’ GAG GAT GGT GGT CAA GGG AC 3’
HV2a
F15
R285
5’ CAC CCT ATT AAC CAC TCA CG 3’
5’ GTT ATG ATG TCT GTG TGG AA 3’
Reference Samples
F15989
R484
5’ CCC AAA GCT AAG ATT CTA AT 3’
5’ TGA GAT TAG TAG TAT GGG AG 3’

HV1a and HV2a amplification reactions were set up for six of the ten unhandled
cartridge casings swabbed by Orlando (2012), along with the reagent blank. Furthermore,
Zymo-Spin™ IV-HRC columns (Zymo Research, Irvine, CA) were tested for the presence of

21

mtDNA; a column was added to a collection tube and centrifuged for 3 min at 8,000 rcf to
remove the liquid that the matrix comes suspended in. Following UV irradiation for 10 min, 50
µL of TE (10 mM Tris [pH 7.5], 1 mM EDTA) was added to the column, and centrifuged at
8,000 rcf for 1 min (the filtrate was denoted T1). The same protocol was repeated with a new
column, but it was exposed to UV for 10 min prior to the first spin and for 5 min after the liquid
was centrifuged through (T2). Both filtrates and a negative control were subjected to HV1a and
HV2a amplification.
Polymerase chain reaction (PCR) was performed in 30 µL volumes and included: 3 µL of
25 mM MgCl2 (Applied Biosystems, Carlsbad, CA), 3 µL of GeneAmp 10X PCR Buffer II
(Applied Biosystems), 3 µL of 2 mM deoxynucleoside 5’-triphosphates, 3 µL of 4 µg/µL bovine
serum albumin (BSA; Fisher Scientific, Waltham, MA), 3 µL of 20 µM forward and reverse
primers, 1 unit of AmpliTaq Gold® polymerase (Applied Biosystems), 1 µL of template DNA,
and 11 µL of water. Six live Orlando (2012) cartridge casing extracts, T1, T2, reagents blanks,
and negative controls were amplified in 10 µl volumes with one third the amount of reagents,
and 1 µL of extract. PCR conditions were: 94˚C for 10 min, 38 cycles of 94˚C, 60˚C, and 72˚C
for 30 s each, and a final extension step of 72˚C for 5 min. Reference samples were amplified
using the same PCR conditions, but for 35 cycles. Five microliters of PCR products were
electrophoresed on a 4% agarose gel. If no band was present, a new 30 µL reaction was made
using 2 µL of template DNA.

Sequencing Genetic Profiles
PCR products of cartridge casing extracts and reference samples were purified using
Diffinity RapidTip® (Diffinity Genomics, Inc., West Henrietta, NY) as per manufacturer’s

22

instructions. Ten microliter Sanger sequencing reactions were prepared using 2.5 µL of
BigDye® Terminator v3.1 Cycle Sequencing master mix (0.875 µL of BDX64 BigDye®
enhancing buffer [MCLAB, San Francisco, CA], 0.125 µL of BigDye® Terminator v3.1 Ready
Reaction Mix [Applied Biosystems], 1.5 µL of 5X Sequencing Buffer [Applied Biosystems]), 1
µL of one of the same primers used to amplify the DNA, and 6.5 µL of the purified template.
For reference samples, 2 µL of purified template and 4.5 µL of distilled water were added to
bring the volume to 10 µL. Sequencing conditions were: 96˚C for 3 min, followed by 30 cycles
of 96˚C for 10 s, 50˚C for 5 s, and 60˚C for 2 min. For R16410 sequences, 2.7 µL of BigDye®
Terminator v3.1 Ready Reaction Mix, 1.3 µL of 5X Sequencing Buffer, 1 µL of F16410, and 5
µL of amplified DNA from casings or 2 µL of amplified reference sample DNA and 3 µL of
distilled water were used. Thermocyling parameters were: 96˚C for 1 min, followed by 25
cycles of 96˚C for 10 s, 50˚C for 5 s, and 60˚C for 4 min.
Two and a half microliters of stop solution (1 µL of 100 mM EDTA [pH 8], 1 µL of 3 M
NaOAC, 0.5 µL 20 mg/mL glycogen), the 10 µL of sequenced products, and 35 µL of chilled
95% ethanol were added to 1.5 mL microcentrifuge tubes. Tubes were vortexed for 10 s,
centrifuged at full speed for 10 min, and the supernatants were removed without disturbing the
pellets. One hundred-eighty microliters of chilled 70% ethanol was added to the tubes,
centrifuged at full speed for 4 min, and the supernatants removed. The 70% ethanol wash was
repeated twice. DNAs were vacuum-dried for 15 min and resuspended in 10 µL of Hi-Di™
Formamide (Applied Biosystems). DNAs were electrophoresed on an AB 3500 Genetic
Analyzer using instrument parameters: oven temperature 60˚C; run time 1020 s; run voltage 19.5
kV; injection time 8 s; injection voltage 1.6 kV; capillary length 50 cm. A runtime of 1400 s was

23

used for reference samples. Sequences were analyzed using Sequencing Analysis Software v5.4
(Applied Biosystems) and were aligned using BioEdit v7.0.9.0 (Hall, 1999).

Analysis of Genetic Profiles, Consistency of Assignments, and Mixtures
Sequences were compared to the Cambridge Reference Sequence (Anderson et al., 1981).
If the Sequencing Analysis Software deciphered the same polymorphism in both the forward and
reverse sequences, the base call was not altered. However, if differing base calls were made, the
electropherograms were analyzed to determine if there was a mixture at that position. The
nomenclature used for base calls is summarized in Table 3.

Table 3. Nomenclature used to make base calls for all sequences.
Base(s)
Nomenclature
Adenine
A
Cytosine
C
Guanine
G
Thymine
T
Adenine + Cytosine
M
Adenine + Guanine
R
Cytosine + Guanine
S
Cytosine + Thymine
Y

Haplotypes of the analysts and the seven volunteers comprised the database for Part A. The
haplotypes from each extract were randomly assigned a number and a blind study was conducted
to determine if the haplotype could be successfully matched back to the loader included in a
database. When a mixture was detected, assignments were made for haplotypes with and
without the polymorphic base. For example, if the haplotype was a 16291T, 73R, 152C, and
263G, an assignment was attempted for haplotype 16291T, 73A (the reference sequence base),

24

152C, and 263G, as well as for haplotype 16291T, 73G, 152C, and 263G. The assignments were
compared to the loader and assessed using the following rules:
1. Consistent-Single: The haplotype was consistent with a single individual and it was the
true loader.
2. Consistent-Multiple: The haplotype was consistent with a group of individuals and the
loader was included in this group.
3. Inconsistent: The haplotype was consistent with either a single individual or a group of
individuals, but the loader was not one of the individuals.
4. Inconclusive: The haplotype was not consistent with any individual.
The presence of a mixture containing the loader’s DNA (loader mixture) was established by
identifying more than one peak at any polymorphic base consistent with the loader. In instances
when the loader’s mtDNA did not contain any polymorphisms, clean regions of the sequences
were inspected for any position(s) with a second peak that was larger than the baseline. Extracts
with polymorphisms consistent with the primary analyst were noted as ‘contaminated’.

Part B
Paper bags and cotton swabs (860-PPC, Puritan Medical Products Co., Guilford, ME)
were autoclaved at 135°C for 45 min, followed by a dry cycle of 1 hr. A facemask and two pairs
of gloves were worn during sample collection. Lab coat, hair net, sleeves, facemask, and two
pairs of gloves were used during all other pre-amplification procedures. Reagents and
consumables were exposed to ultraviolet (UV) light for a minimum of 5 min per side (~2.5
2

J/cm ) in a Spectrolinker XL-1500 UV Crosslinker (Spectronics Corporation, Westbury, NY)

25

prior to use; swabs were placed at an angle to the source of UV light and were rotated 180
degrees after 5 min of exposure.

Obtaining and Decontaminating Cartridge Casings
Remington® .40-caliber, brass cartridges were purchased at local retail stores. Since
analysis of the Orlando (2012) extracts indicated mtDNA was present on unhandled cartridges,
the casings were wiped with ELIMINase® (Decon Laboratories, Inc., Bryn Mawr, PA) as per
manufacturer’s instructions. Residual ELIMINase® was removed by wiping with deionized
water and drying two times. The decontaminated cartridges were separated into sets of six and
placed into paper bags labeled 1 through 38.

Loading of Cartridges
The use of human subjects as loaders of cartridges was approved by the Michigan State
University (MSU) Committee on Research Involving Human Subjects (IRB# 12-770). Prior to
participation, volunteers from MSU and the MSP Sterling Heights (Collection 1) and Northville
(Collection 2) Forensic Science Laboratories signed consent forms. One firearm and one
magazine were used for Collection 1 and one firearm and five magazines were used for
Collection 2. Before Collection 1, the inside of the magazine was wiped with a swab wetted
with water. Volunteers then randomly selected a paper bag and loaded the six cartridges into the
magazine of a handgun. The cartridges were fired and the casings were collected: for Collection
1, all casings handled by one volunteer fell onto netting around the firing tank and were placed
back into their respective paper bags; for Collection 2, they were caught in heat-seal grade plastic
bags held by a metal wire next to the ejection port. The second through fifth spent casings

26

handled by one volunteer were ejected into one bag, while the first and sixth spent casings were
caught in individual bags. The plastic bags were then placed into their respective paper bags.
Two buccal swabs obtained from each volunteer as DNA reference samples were placed
into a culture tube, and labeled with a letter randomly selected from A – XX. The buccal letter
corresponded with the number on the bag containing the cartridges, not the volunteer, thus all
DNAs were deidentified. Spent cartridge casings and buccal swabs were stored at -20°C.

DNA Recovery and Isolation
The casings of three randomly chosen cartridges (L1, L2, and L3) from the
manufacturer’s box were swabbed using the double-swab technique (Sweet et al., 1997); the first
swab used was wetted with approximately 100 μL of digestion buffer (0.1% SDS, 20 mM Tris
[pH 7.5], 50 mM EDTA) and then a dry swab was used. Three of the six casings from each
Collection 1 volunteer were randomly selected and swabbed with separate pairs of swabs
(‘individually swabbed’) (Figure 6). The remaining three casings were swabbed using a single
pair of swabs (‘cumulatively swabbed’). The first, sixth, and a random middle casing from
Collection 2 were swabbed individually and the other three casings were swabbed cumulatively.
Figure 6 also depicts schematics of the design used to collect casings and recover DNAs.

27

a.

b.

Volunteers loaded 6
decontaminated cartridges

c. Volunteers loaded 6 decontaminated cartridges
nd

st

6 casings caught in netting
& transferred to paper bag

1 casing fired
caught in bag

Random 3 Remaining 3
swabbed
swabbed
individually cumulatively

th

2 –5
casings fired
caught in bag

th

6 casing fired
caught in bag

Swabbed
Swabbed
individually Random 1 Remaining 3 individually
swabbed
swabbed
individually cumulatively

Figure 6. a. Cartridge casings individually double-swabbed on the left and cumulatively doubleswabbed on the right. b. Flow chart of the collection of casings and recovery of cells from
Collection 1. c. Flow chart of the collection of casings and recovery of cells from Collection 2.

Swab heads were clipped into 1.5 mL microcentrifuge tubes containing 400 μL of
digestion buffer and 6 μL of proteinase K (20 mg/mL), the tubes were vortexed for 10 s, and
incubated overnight at 55˚C. The swab heads were transferred to spin baskets using forceps and
centrifuged at full speed for 4 min. The swab heads were discarded and the flow-throughs were
added to the original digestion tubes.
DNAs were extracted by adding 500 μL of phenol to the digestion tubes, which were
vortexed for 10 s and centrifuged at full speed for 5 min. The aqueous layers were transferred to
tubes containing 500 μL of chloroform, vortexed for 10 s, and centrifuged at full speed for 5 min.

28

Amicon® Ultra-0.5 mL, 30kDa spin columns (Millipore Corporation, Billerica, MA) were
pretreated by adding 1 μL of Saccharomyces cerevisiae rRNA (10 mg/mL; Alfa Aesar, Ward Hill,
MA) and 499 μL of low TE (10 mM Tris [pH 7.5], 0.1 mM EDTA), centrifuged for 10 min at
14,000 rcf, and the filtrates were discarded. The aqueous layers were transferred to filters,
centrifuged for 10 min at 14,000 rcf, and flow-throughs were discarded. The DNAs were
washed twice using 300 μL of low TE and centrifuging for 10 min at 14,000 rcf. The filters were
inverted into new Amicon® collection tubes and centrifuged for 4 min at 1,000 rcf. Buccal swab
DNAs were extracted in the same manner, except three TE (10 mM Tris [pH 7.5], 1 mM EDTA)
washes were performed. DNAs were stored at -20°C.

mtDNA Amplification, Sequencing, Analysis of Genetic Profiles, Consistency of Assignments,
and Mixtures
Amplification using primer pairs HV1 and HV2 was tested on extract 12C from
Collection 1, prior to amplification of all extracts using HV1a, HV1b, and HV2a. PCR,
sequencing, analysis of genetic profiles, a blind study, and determination of loader mixtures for
all Part B extracts were performed following protocols used in Part A. However, haplotypes of
all individuals in the MSU Forensic Biology Laboratory and the 28 volunteers from Collections
1 and 2 were included in the Part B database. Sequences for inconsistent and inconclusive
assignments from Collections 1 and 2 were re-analyzed in the Sequencing Analysis Software by
activating the mixed base identification (MBI) option set to 30% peak height as recommended by
Nickerson et al. (1997). A blind study was re-performed using the MBI haplotypes.

29

Construction of Consensus Haplotypes
Consensus haplotypes for Collections 1 and 2 were generated using data from the three
individually swabbed casings (Indiv-C), as well as the individually and cumulatively swabbed
casings (All-C). The haplotypes developed with MBI were used for initial inconsistent and
inconclusive assignments. A polymorphism that was present in at least n – 1 of the extracts from
each loader was included in the consensus profile. Since there were only two individually
swabbed casings from bag 33, polymorphisms that were present in both were included in the
Indiv-C haplotype. When a polymorphic position used in the consensus profile contained a
mixture for all of the extracts, the nomenclature for the mixture of bases was kept; if one or more
extracts contained a clean polymorphism, the corresponding non-mixture nomenclature was
given. Haplotypes were then reassigned.

Statistical Analyses
An Analysis of Variance (ANOVA) was performed on STR profiles from Orlando (2012)
extracts, excluding contaminated samples, to determine if there was a difference in the number of
inconsistent alleles among the mtDNA assignments. This was repeated using the frequency of
consistent alleles. Residuals were extracted, and the Shapiro-Wilk test was used to assess
normality. Homogeneity of variances for each group of mtDNA assignments was compared
using Levene's test. If the assumptions of normality and variance were valid, the parametric
ANOVA was used to compare the means of the groups, whereas if the assumptions were violated,
the non-parametric Kruskal-Wallis test was conducted to compare the medians of the groups. A
Pearson's chi-square test for count data or a Fisher’s exact test was performed to determine
whether the swabbing method influenced the consistency of assignments and to establish if there

30

were significant differences between the assignments made. In instances when there were no
differences in assignments between the swab methods and Collections, the data were combined.
A Pearson's chi-square test was also used to determine if there was a difference in the amount of
loader mixtures among the four swabbing categories used for Collection 2. Significance was
identified for all tests at p < 0.05. XLSTAT (Addinsoft, 1995 – 2013) was used for the KruskalWallis test, while all other statistical tests were performed in R (www.r-project.org).

31

RESULTS

Part A
Degradation of DNA Recovered from Spent Cartridge Casings
There was substantial degradation of DNAs isolated from the fired casings at both the
nuclear and mitochondrial levels. Table 4 lists the frequency of STR alleles consistent with the
loader for the 46 Orlando (2012) DNA extracts that were sequenced in this research. The
smallest amplicons (D8, D3, D19, and amelogenin) had the highest frequency of amplification,
followed by a decreasing trend as the loci became larger, with the CSF locus being the sole
exception.

Table 4. Frequency of consistent alleles amplified at each locus illustrating DNA degradation of
DNAs from Orlando (2012) extracts that underwent mtDNA analysis. Amplicons are arranged
from smallest to largest for each dye. The smaller amplicons, on average, had higher frequencies
of amplification compared to the larger amplicons.
Locus
Frequency of Consistent Alleles Per Locus
Blue Dye
2.48%
D8
0.47%
D21
0.14%
D7
1.09%
CSF
Green Dye
3.26%
D3
2.03%
Tho
1.09%
D13
1.09%
D16
0.16%
D2
Yellow Dye
3.19%
D19
1.67%
vWA
1.51%
TPOX
0.67%
D18

32

Table 4 (cont’d)
Red Dye
6.16%
2.31%
0.43%

Amelogenin
D5
FGA

The levels of mtDNA degradation from fired cartridge casings are exemplified using
Orlando (2012) extracts 22D and 28C (Figure 7). The DNAs amplified using HV1b (F16190 –
R16410; 221 bp) and HV2a (F15 – R285; 271 bp). However, they had lower band intensities
using HV1 (F15989 – R16410; 422 bp) and showed no amplification using HV2 (F15 – R484;
470 bp).

a. Neg

Pos
22D 28C
F15989 – R16410
HV1

b.

Neg

Neg

Pos
22D 28C
F16190 – R16410
HV1b

Pos
22D
F15 – R484
HV2

Neg

Pos
22D 28C
F15 – R285
HV2a

28C

Figure 7. Four percent agarose gel electrophoreses showing PCR products of extracts 22D and
28C from Orlando (2012). Primer pairs were: a. HV1 (left), HV1b (center), HV2a (right), and b.
HV2. No bands were present in the negative controls (Neg), and were in the positive controls
(Pos). The extracts amplified using HV1b and HV2a, had decreased amplification using HV1,
and no amplification using HV2.

33

Contamination of Zymo-Spin™ IV-HRC Columns and Cartridges
mtDNA from Zymo-Spin™ IV-HRC column extract T1 amplified using primer pairs
HV1a and HV2a (Figure 8). In contrast, T2 did not result in amplification. Three of the six live
cartridge casings swabbed by Orlando (2012) prior to handling resulted in mtDNA amplification
using both HV1a and HV2a, as did the reagent blank.

Neg

T1
HV1a

T2

Neg

T1
HV2a

T2

Figure 8. Four percent agarose gel electrophoresis showing PCR results from the Zymo-Spin™
IV-HRC column contamination test. No bands were present in the negative controls and filtrate
T2 using HV1a and HV2a, but there were bands for T1.

mtDNA Amplification, Sequencing, and Assignments of Orlando (2012) DNAs
The average amplification rate on the first try of Orlando (2012) extracts was 97.7%
(HV1a, 95.8%; HV1b, 100.0%; HV2a, 98.1%). Forty-two of the 46 extracts had evidence of
mtDNA mixture. Eight individually and one cumulatively swabbed casing DNAs contained all
polymorphisms consistent with the primary investigator from the Orlando (2012) study and thus
were deemed contaminated and were eliminated from the pool of data. Figure 9a shows the
assignments made for the 30 individually swabbed casings; 11 of the assignments were
consistent-single (36.7%), 3 were inconsistent (10.0%), and 16 were inconclusive (53.3%), while
no consistent-multiple assignments were made. From the seven cumulatively swabbed casings,
three were consistent-single (42.9%) and the remaining four were inconclusive (57.1%; Figure

34

9b). Overall, the swabbing method had no significant effect on assignment categories (p =
0.680). There were significantly more consistent-single than inconsistent assignments as well as
inconclusive than inconsistent assignments when data from individually and cumulatively
swabbed casings were combined (p = 0.008 and p = 3.93E-4, respectively; Table 5). However,
there was not a significant difference between the number of consistent-single and inconclusive
assignments (p = 0.304).

a.
Consistent-Single
n = 11

Inconclusive
n = 16

Inconsistent
n=3
b.
Consistent-Single
n=3
Inconclusive
n=4

Figure 9. Assignments made for Orlando (2012) DNA extracts: a. individually swabbed and b.
cumulatively swabbed spent cartridge casings. The majority of assignments for both swab
methods were inconclusive.

35

Table 5. Pearson’s chi-square p-values comparing assignments for Part A individually and
cumulatively swabbed spent cartridge casings after they were combined. The assignment with
the higher frequency is listed first. There were significantly more consistent-single than
inconsistent, and inconclusive than inconsistent assignments (p = 0.008 and 3.93E-4,
respectively).
Assignment Comparison
p-value
Consistent-Single vs. Inconsistent
0.008
Inconclusive vs. Consistent-Single
0.304
Inconclusive vs. Inconsistent
3.93E-4

Table 6 shows that consistent-single mtDNA assignments had, on average, amplification
of 17.3% of STR alleles consistent with the loader, while inconsistent assignments had a
frequency of 15.3%, and inconclusive assignments had 10.4%. Also listed in Table 6 is the
number of inconsistent STR alleles for each DNA extract; consistent-single assignments had an
average of 1.1 inconsistent alleles, inconsistent mtDNA assignments had 2.0, and inconclusive
assignments had 3.8. Although consistent-single mtDNA assignments had the highest frequency
of consistent STR alleles and fewest inconsistent STR alleles, the average frequencies of
consistent alleles and number of inconsistent alleles from the three groups of mtDNA
assignments did not differ significantly (ANOVA, p = 0.175 and Kruskal-Wallis, p = 0.101,
respectively).

36

Table 6. The percentage of STR alleles consistent with the loader, number of alleles
inconsistent with the loader, and mtDNA assignments for Orlando (2012) DNA extracts. Extract
numbers followed by an A indicate cumulatively and letters B – F indicate individually swabbed
casings. Extracts with a consistent-single mtDNA assignment had a higher average frequency of
consistent alleles, compared to inconsistent assignments (17.3% vs. 15.3%, respectively).
Alleles Consistent Alleles Inconsistent
mtDNA
Extract
with Loader
with Loader
Assignment
2A
35.7%
3
Consistent-Single
2B
35.7%
1
Consistent-Single
2D
21.4%
0
Consistent-Single
6A
28.6%
1
Consistent-Single
6B
25.0%
0
Consistent-Single
6C
17.9%
3
Consistent-Single
6D
10.7%
1
Consistent-Single
6E
25.0%
2
Consistent-Single
22C
3.4%
1
Consistent-Single
22D
17.2%
0
Consistent-Single
22E
0.0%
1
Consistent-Single
22F
3.4%
0
Consistent-Single
28A
18.5%
1
Consistent-Single
28E
0.0%
1
Consistent-Single
2C
17.9%
0
Inconsistent
8B
20.7%
6
Inconsistent
28F
7.4%
0
Inconsistent
3A
22.2%
7
Inconclusive
3B
7.4%
4
Inconclusive
3C
0.0%
1
Inconclusive
3D
33.3%
17
Inconclusive
3E
11.1%
1
Inconclusive
3F
7.4%
4
Inconclusive
5A
18.5%
0
Inconclusive
5B
0.0%
2
Inconclusive
5C
0.0%
0
Inconclusive
5D
0.0%
3
Inconclusive
5E
11.1%
3
Inconclusive
5F
11.1%
6
Inconclusive
7C
28.6%
14
Inconclusive
7D
0.0%
0
Inconclusive
8A
13.8%
0
Inconclusive
22A
10.3%
6
Inconclusive
22B
10.3%
2
Inconclusive
28B
7.4%
4
Inconclusive
28C
14.8%
0
Inconclusive
28D
0.0%
2
Inconclusive

37

Table 6 (cont’d)
2E
2F
6F
7A
7B
7E
7F
8C
8D
Average
Average
Average

7.1%
0.0%
39.3%
17.9%
10.7%
17.9%
7.1%
24.1%
13.8%

0
1
8
1
4
3
2
8
5

Contaminated
Contaminated
Contaminated
Contaminated
Contaminated
Contaminated
Contaminated
Contaminated
Contaminated

17.3%
15.3%
10.4%

1.1
2.0
3.8

Consistent-Single
Inconsistent
Inconclusive

Part B
Degradation of DNA Recovered from Spent Cartridge Casings
PCR results from Collection 1 extract 12C exhibited DNA degradation (Figure 10).
There was mtDNA amplification using primer pairs HV1a and HV2a, no amplification using
HV1, and weak amplification using HV2. Given these results, the smaller amplicons were
assayed in subsequent experiments. The average amplification rate on the first attempt was
97.9% (HV1a, 97.3%; HV1b, 99.1%; HV2a, 97.3%). A second PCR with double the amount of
input DNA yielded successful amplification in all other instances.

38

a.

Neg

12A

12B
12C
HV1a

b.

12D

Neg

Neg
12C
HV1

Neg

RB

12A

12B
HV2a

12C

12D

12C
HV2

Figure 10. a. Four percent agarose gel electrophoreses showing PCR results of Collection 1
extract 12C using primer pairs HV1a, HV2a, b. HV1, and HV2. In 8a., PCR results for extracts
12A – D are shown; 12C extracts are boxed in red. No bands were present in negative controls
and the reagent blank (RB). Extract 12C amplified using HV1a and HV2a, had a lower level of
amplification using HV2, and no amplification using HV1.

Contamination of Cartridges and the Magazine
mtDNA from the three live cartridge casings tested in Part B amplified using primer pairs
HV1a and HV2a (L1 – 3; Figure 11). The mtDNA recovered from the magazine prior to
Collection 1 yielded a haplotype of 16126C, 73G, and 263G using the same primer pairs.

39

RB

Neg

L1
HV1a

L2

L3

RB

Neg

L1
HV2a

L2

L3

Figure 11. Four percent gel electrophoresis showing PCR results from live cartridges L1 – 3
using primer pairs HV1a and HV2a. L1 – 3 amplified, while no bands were present in the
reagent blanks or negative controls.

mtDNA Haplotypes and Loader Assignation for Collections 1 and 2
Three volunteers (SS, TT, and XX) from Collection 2 expressed heteroplasmy.
Furthermore, the only difference between volunteer VV (16069T, 16126C, 73G, 185A, 228A,
263G) and XX (16069T, 16126Y, 73G, 185A, 228A, 263G) was a heteroplasmic site at position
16126 for volunteer XX. Therefore, all extracts with a C at position 16126 in addition to the
remaining five polymorphisms were assigned to both volunteers.
Figure 12 illustrates that 7 of 12 (58.3%) individually swabbed casings and 3 of 4
(75.0%) cumulatively swabbed casings from Collection 1 yielded consistent-single mtDNA
assignments. The remaining five assignments for individually swabbed casings were
inconclusive (41.7%) and the fourth cumulatively swabbed extract yielded an inconsistent
assignment (25.0%). The mtDNA assignments were not significantly different between the two
swab methods (p = 0.091).

40

a.

Consistent-Single
n=7
Inconclusive
n=5

b.

Inconsistent
n=1

Consistent-Single
n=3

Figure 12. Assignments made for Collection 1: a. individually swabbed and b. cumulatively
swabbed spent cartridge casings. The majority of assignments for both swab methods were
consistent-single.

One cartridge from bag 33 was missing, resulting in one less individually swabbed sixth
ejected casing for Collection 2 (23 versus 24 for the other categories). Figure 13 shows that
consistent-single assignments comprised 12 (50.0%) of the first, 9 (37.5%) of the random middle,
10 (43.5%) of the sixth, and 12 (50.0%) of the cumulatively swabbed spent casings from
Collection 2. Consistent-multiple assignments amounted to 5 (20.8%) of the first, 3 (12.5%) of
the random middle, 4 (17.4%) of the sixth, and 5 (20.8%) of the cumulatively swabbed casings.

41

Inconsistent assignments were obtained from 2 (8.3%) of the first, 2 (8.3%) of the random
middle, 3 (13.0%) of the sixth, and 1 (4.2%) of the cumulatively swabbed casings. Inconclusive
assignments included 5 (20.8%), 10 (41.7%), 6 (26.1%), and 6 (25.0%), respectively. The
assignments were independent of the swab method (p = 0.883), and remained so after data from
all individually swabbed cartridge casings (first, random middle, and sixth) were combined (p =
0.768). There was no significant difference between assignments made for individually swabbed
casings between Collections 1 and 2 (p = 0.243). Similarly, the assignments made for
cumulatively swabbed casings between the Collections were not significantly different (p =
0.230). The individually swabbed casings from both Collections were combined, and compared
to the combined cumulatively swabbed casings, which also resulted in no difference between
swabbing methods (p = 0.758). Finally, there was no difference when the consistent-single and
consistent-multiple assignments were combined and assignments were compared between the
combined individual and cumulative swab methods (p = 0.557).

42

a. First Spent Casings
Inconclusive
20.8%

b. Random Middle Spent Casings

Consistent-Single
50.0%

Inconclusive
41.7%

Consistent-Single
37.5%

Inconsistent
8.3%
Consistent-Multiple
20.8%

Inconsistent
8.3%

c. Sixth Spent Casings
Inconclusive
26.1%

Consistent-Multiple
12.5%

d. Cumulatively Swabbed Spent Casings
Inconclusive
25.0%

Consistent-Single
43.5%

Consistent-Single
50.0%

Inconsistent
4.2%

Inconsistent
13.0%

Consistent-Multiple
Consistent-Multiple
20.8%
17.4%
Figure 13. Assignments made for spent casings from Collection 2: a. the first (n = 24), b. a random middle (n = 24), c. the sixth (n =
23), and d. three cumulatively swabbed spent cartridge casings (n = 24). When consistent-single and consistent-multiple assignments
were combined, they comprised 50.0% or more of assignments for the four categories.
43

The four assignment categories were not equally obtained (p = 8.51E-9) after data from
individually and cumulatively swabbed casings were combined from Collections 1 and 2. This
was also true when consistent-single and consistent-multiple assignments were combined (p =
7.26E-12). Figure 14 shows the percentage of assignments made for the 111 extracts from
Collections 1 and 2; 70 (63.1%) of the extracts resulted in consistent assignments, 9 (8.1%) were
inconsistent assignments, and 32 (28.8%) were inconclusive. Results displayed in Table 7
indicate there was a significantly greater number of consistent-combined assignments than
inconsistent assignments (p = 6.74E-12), consistent-combined than inconclusive assignments (p
= 1.68E-4), and inconclusive than inconsistent assignments (p = 3.28E-4).

Inconclusive
28.8%

Consistent-Combined
63.1%

Inconsistent
8.1%

Figure 14. Assignments made for all profiles from Collections 1 and 2; consistent-single and
consistent-multiple assignments were combined. The majority of assignments for all Part B
extracts were consistent-combined (63.1%).

44

Table 7. Pearson’s chi-square p-values obtained when comparing assignments made for
combined individually and cumulatively swabbed spent cartridge casings from Collections 1 and
2. The assignment with the higher frequency is listed first. There were significantly more
consistent-combined than inconsistent and inconclusive assignments (6.74E-12 and 1.68E-4,
respectively). There were more inconclusive than inconsistent assignments (3.28E-4).
Assignment Comparison
p-value
Consistent-Combined vs. Inconsistent
6.74E-12
Consistent-Combined vs. Inconclusive
1.68E-4
Inconclusive vs. Inconsistent
3.28E-4

Detection of Mixtures from Collections 1 and 2
Table 8 shows that one of the seven consistent individually swabbed and two of the three
cumulatively swabbed casings from Collection 1 contained mixtures. The assignments made
with and without MBI, which did not resolve any inconsistent or inconclusive assignments made
for Collection 1, are also compared. The inconsistent assignment and two of the five (40.0%)
inconclusive assignments had a mixture with a haplotype consistent with the loader. Seven of
the 16 (43.8%; 1A, 1B, 1C, 1D, 7B, 7CD, and 13C) extracts contained at least one
polymorphism that could not have originated from any of the loaders (Appendix B).

45

Table 8. mtDNA assignments made for haplotypes developed with and without MBI for
Collection 1. The fourth column indicates if the extracts contained a mixture that included a
haplotype consistent with the loader (loader mixture). Extract numbers followed by an A
indicate cumulatively swabbed and letters B – D represent individually swabbed spent casings.
MBI did not alter any inconsistent or inconclusive assignments. Six of the 16 extracts contained
a loader mixture.
mtDNA Assignment mtDNA Assignment
Loader
Extract
without MBI
with MBI
Mixture
7A
Consistent-Single
Yes
7C
Consistent-Single
Yes
12A
Consistent-Single
N/A
12B
Consistent-Single
N/A
12C
Consistent-Single
N/A
12D
Consistent-Single
N/A
13A
Consistent-Single
Yes
13B
Consistent-Single
N/A
13C
Consistent-Single
Indeterminate*
13D
Consistent-Single
N/A
1A
Inconsistent
Inconsistent
Yes
1B
Inconclusive
Inconclusive
Yes
1C
Inconclusive
Inconclusive
Yes
1D
Inconclusive
Inconclusive
No
7B
Inconclusive
Inconclusive
No
7D
Inconclusive
Inconclusive
No
N/A = not applicable
*the six polymorphic positions consistent with the loader did not contain a mixture, but two
additional mixture positions were detected

Eight of the 17 (47.1%) consistent-combined first from Collection 2, 5 of the 12 (41.7%)
random middle, 11 of the 14 (78.6%) sixth, and 8 of the 17 (47.1%) cumulatively swabbed
cartridge casings contained loader mixtures (Table 9). MBI for the 35 inconsistent and
inconclusive assignment from Collection 2 yielded 17 new assignments (bolded). Nine of these
became consistent-multiple. However, seven inconclusive assignments were converted to
inconsistent assignments and one inconsistent assignment became inconclusive. Four (50.0%) of
the extracts with an initial inconsistent assignment and 17 (63.0%) with inconclusive
assignments contained a mixture consistent with the loader, whereas it was not possible to
46

determine if the remaining two extracts contained a loader mixture. Two of the 95 (2.1%; 11C
and 20C) extracts from Collection 2 had at least one polymorphism that could not have
originated from any of the loaders (Appendix B).

Table 9. mtDNA assignments made for haplotypes developed with and without MBI for
Collection 2. The fourth column indicates if the extracts contained a loader mixture. Extract
numbers followed by an A represent cumulatively swabbed spent casings, and individually
swabbed spent casings are designated by B (first), C (random middle), and D (sixth). MBI
altered 17 of the 35 inconsistent and inconclusive assignments. Fifty-three of the 95 extracts
contained a loader mixture.
mtDNA Assignment mtDNA Assignment
Loader
Extract
without MBI
with MBI
Mixture
Yes
8B
Consistent-Single
Yes
8D
Consistent-Single
N/A
9A
Consistent-Multiple
N/A
9B
Consistent-Multiple
N/A
9C
Consistent-Multiple
Yes
9D
Consistent-Multiple
N/A
11A
Consistent-Single
N/A
11B
Consistent-Single
N/A
11D
Consistent-Single
Yes
15A
Consistent-Multiple
Yes
16A
Consistent-Single
N/A
16B
Consistent-Single
N/A
16C
Consistent-Single
N/A
16D
Consistent-Single
N/A
17A
Consistent-Single
N/A
17B
Consistent-Single
N/A
17C
Consistent-Single
Yes
17D
Consistent-Single
N/A
18A
Consistent-Multiple
Yes
18B
Consistent-Multiple
Yes
18D
Consistent-Multiple
N/A
19A
Consistent-Single
Yes
19B
Consistent-Single
Yes
19C
Consistent-Single

47

Table 9 (cont’d)
19D
20A
20B
20D
21A
21B
21C
21D
24A
24C
25B
25C
25D
27A
30B
30C
30D
31A
32A
32B
32C
32D
33A
33B
34A
34B
34C
34D
35B
36A
36B
36C
37A
37B
37C
37D
15C
15D

Yes
N/A
N/A
Yes
Yes
Yes
Yes
Yes
Yes
Yes
Yes
Yes
Yes
Yes
Yes
N/A
Yes
Yes
Yes
N/A
Yes
Yes
N/A
N/A
N/A
N/A
N/A
Yes
Yes
Yes
Yes
N/A
N/A
N/A
N/A
N/A

Consistent-Single
Consistent-Multiple
Consistent-Multiple
Consistent-Multiple
Consistent-Single
Consistent-Single
Consistent-Single
Consistent-Single
Consistent-Single
Consistent-Single
Consistent-Multiple
Consistent-Multiple
Consistent-Multiple
Consistent-Single
Consistent-Single
Consistent-Single
Consistent-Single
Consistent-Single
Consistent-Single
Consistent-Single
Consistent-Single
Consistent-Single
Consistent-Single
Consistent-Single
Consistent-Single
Consistent-Single
Consistent-Single
Consistent-Single
Consistent-Single
Consistent-Multiple
Consistent-Multiple
Consistent-Multiple
Consistent-Single
Consistent-Single
Consistent-Single
Consistent-Single
Inconsistent
Inconsistent

Inconsistent
Consistent-Multiple

48

No
Yes

Table 9 (cont’d)
27B
27C
27D
31B
31D
38A

Inconsistent
Inconsistent
Inconsistent
Inconsistent
Inconsistent
Inconsistent

Inconsistent
Consistent-Multiple
Inconsistent
Consistent-Multiple
Inconclusive
Inconsistent

No
Yes
Yes
Yes
Indeterminate*
No

No
8A
Inconclusive
Inconclusive
No
8C
Inconclusive
Inconclusive
Yes
11C
Inconclusive
Inconclusive
No
15B
Inconclusive
Inconsistent
Yes
18C
Inconclusive
Inconclusive
Yes
20C
Inconclusive
Inconclusive
No
22A
Inconclusive
Inconclusive
No
22B
Inconclusive
Inconsistent
No
22C
Inconclusive
Inconclusive
No
22D
Inconclusive
Inconsistent
Yes
23A
Inconclusive
Inconclusive
Yes
23B
Inconclusive
Consistent-Multiple
Yes
23C
Inconclusive
Inconclusive
Yes
23D
Inconclusive
Inconclusive
Yes
24B
Inconclusive
Inconclusive
Yes
24D
Inconclusive
Consistent-Multiple
No
25A
Inconclusive
Inconclusive
Yes
30A
Inconclusive
Inconsistent
Yes
31C
Inconclusive
Consistent-Multiple
Yes
33C
Inconclusive
Consistent-Multiple
Yes
35A
Inconclusive
Consistent-Multiple
Indeterminate**
35C
Inconclusive
Inconclusive
Yes
35D
Inconclusive
Consistent-Multiple
Yes
36D
Inconclusive
Inconsistent
Yes
38B
Inconclusive
Inconsistent
Yes
38C
Inconclusive
Inconclusive
No
38D
Inconclusive
Inconsistent
N/A = not applicable
*not possible to determine if one of the polymorphisms consistent with the loader was above
background
**contained four of five polymorphisms consistent with the loader

49

Table 10 summarizes the mixture frequencies of consistent-combined assignments from
Collection 2, as well as the percentage of extracts containing a loader mixture: 8 of the 17
(47.1%) first, 5 of the 12 (41.7%) random middle, 11 of the 14 (78.6%) sixth, and 8 of the 17
(47.1%) cumulatively swabbed casings had a mixture. Adding in inconclusive and inconsistent
assignments, 12 of the 24 (50.0%) first, 13 of the 24 (54.2%) random middle, 17 of the 23
(73.9%) sixth, and 11 of the 24 (45.8%) cumulatively swabbed casings contained a mtDNA
mixture consistent with the corresponding loader’s DNA. The frequency of loader mixture was
not statistically different among the four groups (p = 0.223).

Table 10. Percentage of consistent-combined assignments, inconsistent and inconclusive
assignments, mixture, and loader mixture made for the four Collection 2 categories. The second
and fourth columns summarize the percentage of consistent-combined, and inconsistent and
inconclusive assignemnts made, respectively. The third and fifth columns indicate the
percentage of mixture and loader mixture, respectively, for the two combined assignment types.
The sixth column contains the percentage of the four categories with a loader mixture. The sixth
spent casings had the highest frequency of loader mixture (73.9%), while cumulatively swabbed
casings had the lowest (45.8%).
Consistent-Combined Inconsistent & Inconclusive
Loader
Total
Mixture
Total
Loader Mixture
Mixture
First
Random Middle
Sixth
Cumulatively

70.8%
50.0%
60.9%
70.8%

47.1%
41.7%
78.6%
47.1%

29.2%
50.0%
39.1%
29.2%

57.1%
66.7%
66.7%
42.9%

50.0%
54.2%
73.9%
45.8%

Assignments for Collections 1 and 2
A summary of the assignments made for individually, cumulatively, and all casings from
Collections 1 and 2, as well as the percentage of assignments obtained with the consensus
methods (All-C and Indiv-C) are shown in Table 11. Of the 83 individually swabbed casings, 50
(60.2%) yielded consistent-combined assignments, 7 (8.4%) were inconsistent, and 26 (31.3%)

50

were inconclusive. From the 28 cumulatively swabbed casings, 20 (71.4), 2 (7.1%), and 6
(21.4%) respective assignments were made. The 28 Indiv-C haplotypes yielded 19 (67.9%)
consistent-combined, 4 (14.3%) inconsistent, and 5 (17.9%) inconclusive assignments. All-C
resulted in 19 (67.9%) consistent-combined assignments, 3 (10.7%) inconsistent, and 6 (21.4%)
inconclusive. The assignments were not significantly different among the five methods (p =
0.858).

Table 11. Percentage of consistent-combined, inconsistent, and inconclusive assignments made
from individually, cumulatively, and all swabbed spent casings, as well as All-C and Indiv-C
from Part B. Cumulatively swabbed casings had the highest frequency of consistent-combined
assignments (71.4%), while individually swabbed casings had the lowest frequency (60.2%).
Both Indiv-C and All-C resulted in 67.9% consistent-combined assignments.
Consistent-Combined
Inconsistent
Inconclusive
60.2%
8.4%
31.3%
Individually Swabbed
71.4%
7.1%
21.4%
Cumulatively Swabbed
63.1%
8.1%
28.8%
All Swabbed
67.9%
14.3%
17.9%
Indiv-C
67.9%
10.7%
21.4%
All-C

51

DISCUSSION

Spent cartridge casings found at a crime scene may harbor valuable genetic information
that can be used by law enforcement to connect a perpetrator to a crime. STR analysis has
become the method of choice for DNA identification in forensic science because the genotypic
information obtained has exceptional discriminating capabilities. However, STR analysis of
touch DNA and compromised biological samples is not as successful as analysis of high
molecular weight DNA. Despite attempts to overcome challenges associated with touch DNA
recovered from spent casings, such analysis has met with minimal success, even in controlled
environments (Spear et al., 2005; Horsman-Hall et al., 2009; Branch, 2010; Dieltjes et al., 2011;
Orlando, 2012). In contrast, mtDNA analysis from spent cartridge casings has not been
previously investigated; thus the primary objective of the research presented here was to
determine if the unique traits of mtDNA would allow forensic biologists to successfully develop
the haplotype of a loader.
Humans shed many thousands of skin cells a day (Roberts and Marks, 1980) and as a
result, DNA is deposited through physical contact during the normal use of objects (van Oorshot
and Jones, 1997). DNA has been recovered from items such as steering wheels (Pizzamiglio et
al., 2004; Brevnov et al., 2009), drinking containers (Abaz et al., 2002; Brevnov et al., 2009),
various types of handles (Ladd et al., 1999), firearms (Horsman-Hall et al., 2009; Richert, 2011;
Nunn, 2013), backpacks (Hoffmann et al., 2012), and improvised explosive devices (Esslinger et
al., 2004). Horsman-Hall et al. (2009) hypothesized that full STR profiles were unlikely to be
obtained from DNA left on spent cartridge casings due to a combination of low levels of
deposition, PCR inhibition from gunshot residue, and DNA degradation caused by temperatures

52

reaching upwards of 1800°C in the chamber of the firearm. Similar to the aforementioned touch
DNA studies, the authors recovered DNA from live cartridge casings handled for 30 s (more than
the time needed to load a cartridge). However, like other handled items, the amount of DNA
deposited onto cartridge casings during the simulated loading process was low—14.4 pg/µL—
less than the concentration recommended for STR analysis.
PCR inhibitors, likely in gunshot residue, co-extract with DNA, providing an added
challenge in analyzing recovered DNAs from spent casings. Horsman-Hall et al. (2009) detected
PCR inhibition from 11% of shotgun shells and Orlando (2012) encountered inhibition during
DNA quantification prior to additional purification using Zymo-Spin™ IV-HRC columns.
Owing to this, BSA was added to PCR assays in this study. BSA has been shown to help
overcome PCR inhibition (Kreader, 1996; Eilert and Foran, 2009), likely due to inhibitor binding
capabilities (Dufour and Dangles, 2005; Fasano et al., 2005). BSA is a common additive in
some STR kits, such as Identifiler®, Identifiler® Plus, MiniFiler™ (Applied Biosystems, 2012),
and PowerPlex® 16 System (Promega, 2013) and is included in the standard operating procedure
for amplifications of degraded skeletal remains at The Armed Forces DNA Identification
Laboratory (AFDIL), which uses 0.625 μg/μL of BSA in PCR (Edson et al., 2004). The mtDNA
PCR containing BSA in this study did not appear to be inhibited.
DNA degradation is another obstacle faced by forensic biologists. Various
environmental factors, such as sunlight, humidity, and heat lead to the breakdown of DNA. The
barrel of an M16A1 rifle after firing 180 rounds over the span of 45 s has been documented to
reach 889°C (Elbe, 1975), although this is only half the 1800°C value purported by HorsmanHall et al. (2009). DNA begins to degrade at 130°C under arid conditions, with full degradation
documented at 190°C (Karni et al., 2013). Using a thermal imaging camera, Gashi et al. (2010)

53

reported the temperature of 9 mm brass cartridge casings reaches a maximum of 63°C during
firing. Therefore, the outer surfaces of casings may not get hot enough to degrade the DNA,
particularly the first few that are fired. DNA is also degraded during keratinization (Kita et al.,
2008), thus it is expected that DNA from cells shed onto touched items is degraded. The lack of
amplification of loader alleles at the larger loci in Part A indicated the samples were degraded
(Bender et al., 2004). Furthermore, the MiniFiler™ STR kit, specifically designed to target
small amplicons (70 – 283 bp in length) (Applied Biosystems, 2012c), has been tested on DNA
recovered from spent casings, but full STR profiles were rarely obtained (Horsman-Hall et al.,
2009; Branch, 2010). mtDNA degradation was also encountered in this study; larger mtDNA
amplicons (~450 bp) either resulted in minimal or no amplification, whereas ~250 bp amplicons
almost always amplified. As a result, overlapping mtDNA amplicons were successfully used to
generate mtDNA profiles. Since forensic laboratories strive to maximize casework turn-around
rates, it would be less time consuming to initially target smaller overlapping amplicons that
amplify more reliably (Gabriel et al., 2001), as opposed to first attempting to amplify larger
amplicons and later resorting to smaller amplicons after amplification failure.
Prior to this study, it was not known whether mtDNA results could be obtained from
spent casings. Therefore, Orlando (2012) DNA extracts, which were available at the MSU
Forensic Biology Laboratory, were processed in Part A to answer the primary question: can
mtDNA haplotypes consistent with the loader be developed from DNA left behind on cartridge
casings during loading? It was hypothesized that loaders’ cells would be deposited onto a
cartridge casings during loading, and as a result, their mtDNA profiles would be obtained from
the spent casings. This was the case for 14 of the 37 haplotypes developed from DNAs
recovered in Part A. Only one of the four extracts that did not contain a mtDNA mixture

54

resulted in a haplotype consistent with the loader, indicating the DNA extractions from three of
the casings had become contaminated. The Zymo-Spin™ IV-HRC columns used to remove
inhibitors were not UV irradiated prior to use (Orlando, 2012), and given mtDNA was recovered
from a column, the possibility that the extracts were contaminated via the columns cannot be
ruled out. However, the haplotypes potentially resulting from contamination were not consistent,
which is contrary to a single contamination source. In this regard it is possible that there were
multiple sources of DNA that contaminated the Zymo™ columns. To determine if the columns
did in fact contribute to the contamination, mtDNAs recovered from several columns would need
to be sequenced and haplotypes compared to the casing extracts.
The second question in this study, how do mtDNA results from spent casings compare to
STR results?, was next examined by comparing mtDNA data to the Orlando (2012) Identifiler®
Plus data. mtDNA haplotypes consistent with the loader were generated for 14 of the extracts,
whereas none yielded full STR profiles. Furthermore, less than a fifth of the alleles from the
loader amplified. These differences likely resulted from the heightened sensitivity of mtDNA
analysis. In addition, possible correlations between mtDNA mixtures and non-loader STR
alleles were examined. Since almost all of the mtDNA profiles were mixtures, it is not
surprising that STR alleles inconsistent with the loader also amplified. However, 11 of the
extracts that contained a mtDNA mixture did not have any inconsistent STR alleles. Therefore,
the heightened sensitivity of mtDNA analysis resulted in the detection of non-loader mtDNA
from some extracts, while STR analysis did not. Nevertheless, it would still be beneficial to
perform mtDNA analysis when casings are recovered from crime scenes, since haplotypes were
developed from all casings.

55

Data collected at the MSU Forensic Biology Laboratory subsequent to the Orlando
(2012) study demonstrated an increase in DNA yields when Amicon® columns were pretreated
with RNA. Consequently, the loss of the loaders’ DNAs during purification might have caused a
decrease in the recovery of loader alleles. Furthermore, as discussed above, contamination from
the Zymo™ columns could have resulted in inconsistent alleles. Owing to the amount of DNA
contamination in the Orlando (2012) DNA extracts, a new round of firing was undertaken (Part
B) and precautions were made to minimize the risk of contamination: the Zymo™ columns
were not used, the primary analyst wore a hairnet, and gloves were periodically wiped with 70%
ethanol during pre-PCR steps to remove any DNAs that may have been on the gloves.
Despite the increased precautions taken to avoid contamination, the netting used to catch
the fired casings still represented a possible source of contamination in Collection 1 of Part B. If
exogenous cells on the netting contaminated the casings, a lower frequency of consistent
assignments and/or more mixtures would be expected from Collection 1 compared to Collection
2 when the netting was not used. However, Collections 1 and 2 had similar levels of consistent
assignments (62.5% and 63.2%, respectively), while a greater percentage of DNA extracts from
Collection 2 contained a loader mixture (37.5% vs. 55.8%). The lower frequency of loader
mixture may be due to chance, given that the sample size was small (n = 16). Interestingly,
43.8% of Collection 1 DNAs contained at least one mtDNA polymorphism that could not have
been contributed by any of the loaders, compared to only 2.1% from Collection 2, which
suggests that exogenous cells on the netting likely contributed to contamination. However, this
number may be skewed given that Collection 2 contained many more loader haplotypes, thus any
contaminating mtDNA polymorphisms would be less likely to be inconsistent with all loaders.
Ultimately, since there were similar levels of consistent assignments between the Collections and

56

a greater percentage of loader mixture from Collection 2 extracts, the plastic bags did not appear
to provide an advantage in preventing contamination, and the source of contamination was likely
not due to the way the casings were caught. Of course, since casings ejected at a crime scene fall
onto less than pristine surfaces, the use of the nets likely comes closer to simulating casework
conditions than collecting casings in individual bags.
Horsman-Hall et al. (2009) found that alleles consistent with previous loaders (average of
3 alleles) amplified using MiniFiler™, when post-firing swabs of the breechface, ejection port,
and chamber of firearms were tested. If DNA is already present on the surfaces of the firearms,
it is possible that casings could become contaminated when they come into contact with them,
resulting in mixtures. Moreover, gun sharing is becoming more common among gang members
(Freund, 2012), and the possibility of mixtures is likely increasing. Therefore, the use of one
firearm for each Collection in this study mimicked a scenario increasingly encountered in
firearm crimes. If contamination of the casings from the surfaces of a firearm does occur,
haplotypes consistent with the previous loader(s) would be detected when sharing guns. Taking
note of the sequence that cartridges are loaded into multiple magazines and are fired would allow
researchers to examine any such correlation in more detail; in an effort to maintain anonymity in
this study, however, the order that the magazines were used was not recorded.
van Oorschot and Jones (1997) reported that DNA can be transferred from one
individual to another, for example when shaking hands. Secondary transfer of exogenous DNA
on one’s hand to a touched item has been documented (Lowe et al., 2002; Goray et al., 2010),
and could also have contributed to contamination in this study. For example, Daly et al. (2012)
autoclaved 300 wood, glass, and fabric items, followed by UV irradiation, and had volunteers
firmly hold them for 60 s. While the majority of handled items did not contain a STR mixture,

57

approximately 10% did, which indicated exogenous DNA(s) that was present on some of the
volunteers’ hands was transferred onto the items. Volunteers in the present study were not asked
to wash their hands prior to loading the cartridges, thus foreign DNA picked up immediately
before their participation could have been transferred onto the casings. Furthermore, a large
majority of the volunteers from Collection 2 were married and DNA from intimate partners has
been found to persist on individuals for longer periods of time, likely because there is a greater
quantity of DNA deposited (Matte et al., 2012). Therefore, the DNA from an intimate partner
could also have been deposited during loading of the cartridges. In future studies, it may be
useful to have volunteers submit buccal swabs from their intimate partners and their haplotypes
compared to those recovered from the casings.
Individual bags were used to separately collect the first and sixth spent casings in
Collection 2. Results from these casings were used to answer the third study question: does the
order cartridges are loaded/fired influence mtDNA results? There is immense variation in the
amount of DNA deposited onto touch samples among individuals (Alessandrini et al., 2003;
Raymond et al., 2009; Daly et al., 2012), as well as in the quantity an individual deposits each
time he/she touches an item (Thomasma and Foran, 2013). Goray et al. (2010) found that an
increase in applied pressure correlated with an increase in the amount of DNA transferred.
Therefore, it was hypothesized that in this study, the greatest amount of the loader’s DNA would
be deposited onto the last loaded cartridges (first spent casings), which require the most force to
load into the magazine. However, no significant difference between the assignments made for
the first and last loaded cartridges was found, although the first loaded (last fired) contained a
lower percentage of consistent assignments (60.9% vs. 70.8%). Even though there was no
statistical difference in the percentage of loader mixture among the four Collection 2 categories

58

(p = 0.223), a higher frequency of loader mixture resulted from the first loaded cartridges
compared to the random middle and sixth loaded (73.9%, 54.2%, and 50.0%, respectively),
indicating contamination decreased as cartridges were loaded. It is possible that most of the
exogenous cells on the loaders’ hands were deposited onto the first cartridge loaded, and as a
result, there was decreasing contamination for subsequent cartridge casings. Furthermore, even
though the haplotype recovered from the magazine used for Collection 1 was not detected in the
cartridge swabbing extracts, a magazine cannot be ruled out as a source of DNA contamination.
Approximately the top fifth of the magazine used for Collection 1 was swabbed, whereas the first
loaded cartridge in a series of six is pushed further down into the magazine. Since this cartridge
is exposed to the most surface area of the magazine, it is possible that cells from a previous
loader may be picked up at a greater frequency compared to the other cartridges. However, due
to the variability of touch DNA, it is important to note that with the data obtained in this study,
no definitive conclusions can be drawn regarding the source(s) of contamination.
An additional goal of this study was to identify a DNA recovery method that would yield
the most consistent mtDNA profiles. Two swabbing methods were compared, which addressed
the fourth study question: are there differences in mtDNA results from individually and
cumulatively double-swabbed spent casings? The double-swab method (Sweet et al., 1997) is an
efficient way to recover DNA from surfaces; the wet swab is used to rehydrate the forensic
sample, which likely loosens the cells from the substrate and causes the cells to cling to the swab.
When a dry swab is swirled over the same surface, more of the rehydrated cells are collected.
What has not been studied is whether presumably related evidence should be double-swabbed
individually or cumulatively, as there are potential advantages and disadvantages to both.
Orlando (2012) found that cumulatively swabbing cartridge casings increased the average

59

quantity of DNA recovered. In addition, it is more cost and time effective because reagents and
supplies are only needed for one extract. However, DNA loss has been documented with
cumulatively swabbing, wherein cells that adhered onto the swab from one surface can be
deposited onto the next swabbed surface (Hebda et al., in press). Furthermore, if multiple
sources of DNA are present on the different surfaces, there is a greater chance that mixtures will
result. Similar to Orlando (2012), a greater percentage of profiles from cumulatively swabbed
casings yielded haplotypes consistent with the loader (71.4% vs. 60.2% for individual
swabbings), though the difference was not statistically significant. Cumulatively swabbing also
had a slightly lower percentage of loader mixtures (50.0% vs. 53.6%). Melton et al. (2012)
suggested that a polymorphic variant comprising less than 10% of the total mtDNA in a mixture
could go undetected. While there was a higher possibility of picking up DNAs that originated
from multiple sources when cumulatively swabbing cartridge casings, it seems that 90% or more
of the DNA recovered from 20 of 28 swabs originated from the loader. As a result, higher loader
DNA recovery using cumulatively swabbing likely swamped out any contaminant DNA.
There were instances of mixture where the ratios of the peaks at a polymorphic position
were dissimilar in the forward and reverse sequences from the same PCR product. For example,
extract 13C from Collection 1 had a 3:1 ratio of A:G at position 153 in the F15 sequence, but a
2:3 ratio in the R285 sequence. Since the ratios were not consistent, mixture nomenclature (e.g.
R for a A/G mixture) was used. Inconsistently and inconclusively assigned haplotypes were reanalyzed using MBI to determine if the analysis method would aid in detecting the loader’s
haplotype when the loader’s DNA was not as abundant as other DNA variants. While 25.7% of
MBI profiles from Collection 2 resulted in consistent-multiple assignments (‘best-case scenario’),
supporting the theory that the loaders’ mtDNAs were not as abundant as the contaminant

60

mtDNA(s), seven of the inconclusive haplotypes (25.9%) resulted in inconsistent assignments
(‘worst-case scenario’). Until larger sample sizes are obtained, it is not recommended to use
MBI when analyzing DNA recovered from casings. Electropherograms of consistent and
inconsistent assignments were also examined manually to determine if they contained loader
mixtures, to identify differences between Collections 1 and 2, and to examine the order
cartridges were loaded. Investigating the possibility of a suspect’s inclusion when there were
inconsistent assignments could result in useful information that may otherwise be overlooked.
However, this may only be useful for investigative purposes because statistics that attempt to put
an inclusion interpreted from a mtDNA mixture are currently not admissible in court (Melton et
al., 2012).
The last question addressed in this study was: is it advantageous to use consensus
profiling? Consensus profiling has the potential to filter out inconsistent polymorphisms based
on the premise that the loaders’ mtDNAs should be recovered from more spent casings than
would contaminating DNA. In other words, seeing the same polymorphism multiple times
develops confidence that it originated from the loader. Both consensus methods (Indiv-C and
All-C) yielded more consistent-combined assignments than individually swabbing (67.9% and
67.9% vs. 60.2%), but neither was as effective as cumulatively swabbing (71.4%; though these
differences were not significantly different). Consensus profiling did not outperform
cumulatively swabbing, partly because three sets of individually swabbed cartridge casings (15,
18, and 31; Appendix B) exhibited shared non-loader polymorphisms that were not detected
from the corresponding cumulatively swabbed casings. Since a polymorphism that showed up
two out of three times for Indiv-C and three out of four for All-C haplotypes were included in the
consensus profile, the detection of the contaminant DNA from the individual extracts influenced

61

overall results. However, because non-loader polymorphisms were detected more frequently in a
minority of the individually swabbed casings, consensus profiling can be used to ‘weed out’
some of the inconsistencies when multiple casings are individually swabbed.

62

CONCLUSION

The results of this study establish the viability of mtDNA analysis as a method to
generate valuable genetic information from spent cartridge casings. Despite the degraded nature
of the DNA, it was possible to develop haplotypes with the use of overlapping amplicons.
Owing to the sensitivity of mtDNA analysis, haplotypes were obtained from all individually and
cumulatively double-swabbed casings, approximately two-thirds of which were consistent with
the loader. However, mixtures were also obtained, which were most prevalent from the first
loaded cartridges and the least from the cumulatively swabbed casings. Consensus profiles were
used to filter out non-recurring polymorphisms, and resulted in a larger percentage of consistent
haplotypes compared to the individually swabbed spent casings. Nonetheless, cumulative
swabbing remained the superior DNA recovery method, as it yielded the highest frequency of
consistent haplotypes.
The research presented leads to several other sets of experiments that could be performed.
It would be valuable to analyze mtDNA recovered from various calibers and casing metals to
establish if a similar level of consistency exists with the different ammunition. Since it is
possible that the use of a single gun for each Collection contributed to contamination, it would
also be helpful to examine if firing cartridges from a gun handled and loaded by the owner
results in fewer mixtures. As more sensitive DNA analysis techniques are developed, it would
also be worthwhile to investigate various DNA recovery and extraction method combinations,
coupled with STR amplification using different kits. This would aid in optimizing a strategy for
STR analysis from spent cartridge casings.
mtDNA haplotypes consistent with the loader were successfully recovered from casings

63

that were caught in a non-decontaminated net following firing. Since the loaders’ mtDNA
profiles were obtained when using non-cleaned firearms and magazines, it seems likely that the
loaders’ mtDNA profiles could be successfully developed from casework samples. Furthermore,
the ability to generate mtDNA haplotypes from individual casings provides an added benefit
when only one casing is recovered at a crime scene or when investigators want to compare the
DNA profiles obtained from multiple cartridge casings. However, when possible, fired casings
should be cumulatively swabbed to maximize the probability of recovering the loaders’
haplotypes without the detection of mixture.

64

APPENDICES

65

APPENDIX A

mtDNA Profiles from Orlando (2012)
Tables of mtDNA profiles recovered from Orlando (2012) spent cartridge casings;
Extract A = cumulatively swabbed spent casings;
Extracts B – F = individually swabbed spent casings;
Ø = no polymorphisms;
N/A = sequencing not attempted;
Yellow highlight = polymorphism consistent with the loader;
Note: all haplotypes contain a 263G.

Table A1. mtDNA profiles obtained from spent cartridge casings loaded by volunteer H.
mtDNA
Extract
HV1a
HV1b
HV2a
Assignment
16162G, 16189C,
H
16209C
Ø
73G
16256T, 16270T,
3A
16192T
16272G
73G
Inconclusive
3B
16124C, 16148C
16304C, 16309G
73G
Inconclusive
3C
16129A
16362C
Ø
Inconclusive
73G, 199C, 203R,
3D
16189C
16311C
204C, 250C
Inconclusive
3E
Ø
16362C
73G, 146C, 200G
Inconclusive
16234T, 16271C,
3F
16093C
16362C
73G, 146C, 152C
Inconclusive

Table A2. mtDNA profiles obtained from spent cartridge casings loaded by volunteer K.
mtDNA
Extract
HV1a
HV1b
HV2a
Assignment
16069T, 16126C,
K
16145A, 16172C
16261T
73G, 242T
16069T, 16126C,
28A
16145A, 16172C
16261T
73G, 242T
Consistent-Single
28B
Ø
16311Y
73G, 146C, 152C
Inconclusive
28C
Ø
16311Y
73G, 146C, 152C
Inconclusive
28D
Ø
16311C
73G, 152Y
Inconclusive
16069Y, 16126C,
28E
16145A, 16172Y
16261T
73G, 146Y, 242T
Consistent-Single
28F
Ø
Ø
Ø
Inconsistent

66

Table A3. mtDNA profiles obtained from spent cartridge casings loaded by volunteer O.
mtDNA
Extract
HV1a
HV1b
HV2a
Assignment
16256T, 16292T,
O
16126C
16294T
73G, 146C, 152C
22A
Ø
16256T, 16270T
73G
Inconclusive
22B
16126C
Ø
73G, 152C
Inconclusive
22C
16126C
N/A
73G, 146C, 152C
Consistent-Single
16256T, 16292C,
22D
N/A
16294T
73G, 146Y, 152Y
Consistent-Single
16256N/A, 16292C,
22E
16126C
16294T
73G, 146C, 152C
Consistent-Single
16256N/A, 16292C,
22F
16126C
16294T
73G, 146C, 152C
Consistent-Single

Table A4. mtDNA profiles obtained from spent cartridge casings loaded by volunteer P.
mtDNA
Extract
HV1a
HV1b
HV2a
Assignment
P
16196A
Ø
Ø
5A
Ø
N/A
73R, 152C, 200A
Inconclusive
73G, 146C, 152C,
5B
Ø
N/A
200A
Inconclusive
5C
N/A
N/A
200A
Inconclusive
73G, 146C, 152C,
5D
Ø
N/A
200A
Inconclusive
73G, 146Y, 152C,
5E
Ø
N/A
200A
Inconclusive
5F
16094C
N/A
N/A
Inconclusive

Table A5. mtDNA profiles obtained from spent cartridge casings loaded by volunteer R.
mtDNA
Extract
HV1a
HV1b
HV2a
Assignment
R
Ø
Ø
146C, 195C
8A
Ø
N/A
73G
Inconclusive
8B
Ø
N/A
Ø
Inconsistent
8C
Ø
N/A
73G, 146C, 152C
Contaminated
8D
Ø
N/A
73G, 146C, 152C
Contaminated

67

Table A6. mtDNA profiles obtained from spent cartridge casings loaded by volunteer T.
mtDNA
Extract
HV1a
HV1b
HV2a
Assignment
T
16092C
Ø
Ø
2A
16092C
Ø
Ø
Consistent-Single
2B
16092C
Ø
Ø
Consistent-Single
2C
Ø
Ø
Ø
Inconsistent
2D
16092C
Ø
Ø
Consistent-Single
2E
Ø
16311C, 16362C
73G, 146C, 152C
Contaminated
2F
16092C
16311Y, 16362C
73G, 146C, 152C
Contaminated

Table A7. mtDNA profiles obtained from spent cartridge casings loaded by volunteer U.
mtDNA
Extract
HV1a
HV1b
HV2a
Assignment
U
Ø
Ø
Ø
7A
Ø
N/A
73R, 146C, 152C
Contaminated
7B
Ø
N/A
73G, 146C, 152C
Contaminated
7C
16410C, 16189C
N/A
73G, 210G
Inconclusive
7D
16140C, 16189C
N/A
146C
Inconclusive
7E
Ø
N/A
73G, 146C, 152C
Contaminated
7F
Ø
N/A
73G, 146C, 152C
Contaminated

Table A8. mtDNA profiles obtained from spent cartridge casings loaded by volunteer DD.
mtDNA
Extract
HV1a
HV1b
HV2a
Assignment
DD
Ø
16216G
72C
6A
Ø
N/A
72C
Consistent-Single
6B
Ø
N/A
72C
Consistent-Single
6C
Ø
N/A
72C
Consistent-Single
6D
Ø
N/A
72C
Consistent-Single
6E
Ø
N/A
72C
Consistent-Single
6F
Ø
N/A
73G, 146C, 152C
Contaminated

68

APPENDIX B

mtDNA Profiles from Collections 1 and 2
Tables of mtDNA profiles recovered from Collections 1 and 2 spent cartridge casings;
Extract A = cumulatively swabbed spent casings;
Extract B = individually swabbed first spent casing;
Extract C = individually swabbed random middle spent casing;
Extract D = individually swabbed sixth spent casing;
Ø = no polymorphisms;
Yellow highlight = polymorphism consistent with the loader;
MBI = mixed base identification: haplotypes developed with the mixed base identification option
set to 30% peak height during analysis;
All-C = consensus haplotype developed from individually and cumulatively swabbed spent
cartridge casings;
Indiv-C = consensus haplotype developed from individually swabbed spent cartridge casings;
Note: all haplotypes contain a 263G.

Table B1. mtDNA profiles obtained from spent cartridge casings loaded by volunteer M during
Collection 1.
mtDNA
Extract
Haplotype
Assignment
M
16126C, 16294T, 16304C, 73G
12A
16126C, 16294T, 16304C, 73G
Consistent-Single
12B
16126C, 16294T, 16304C, 73G
Consistent-Single
12C
16126C, 16294T, 16304C, 73G
Consistent-Single
12D
16126C, 16294T, 16304C, 73G
Consistent-Single
All-C
16126C, 16294T, 16304C, 73G
Consistent-Single
Indiv-C
16126C, 16294T, 16304C, 73G
Consistent-Single

69

Table B2. mtDNA profiles obtained from spent cartridge casings loaded by volunteer S during
Collection 1.
mtDNA
Extract
Haplotype
Assignment
S
16126C, 16294T, 16296T, 16304C, 73G, 103A
13A
16126C, 16294T, 16296T, 16304C, 73G, 103A
Consistent-Single
13B
16126C, 16294T, 16296T, 16304C, 73G, 103A
Consistent-Single
13C
16126C, 16294T, 16296T, 16304C, 73G, 103R, 153R, 195Y
Consistent-Single
13D
16126C, 16294T, 16296T, 16304C, 73G, 103A
Consistent-Single
All-C
16126C, 16294T, 16296T, 16304C, 73G, 103A
Consistent-Single
Indiv-C
16126C, 16294T, 16296T, 16304C, 73G, 103A
Consistent-Single

Table B3. mtDNA profiles obtained from spent cartridge casings loaded by volunteer HH
during Collection 1.
mtDNA
Extract
Haplotype
Assignment
HH
16291T, 16304C, 146C
7A
16291T, 16304C, 146C
Consistent-Single
7B
16270T, 16296T, 146Y, 150Y, 200R
Inconclusive
7B-MBI
16270T, 16296Y, 73R, 146Y, 150Y, 200R, 236Y
Inconclusive
7C
16291Y, 16294Y, 16304Y, 146C
Consistent-Single
7D
16126C, 73G, 152C
Inconclusive
7D-MBI
16126C, 16278Y, 16296Y, 73G, 152Y
Inconclusive
All-C
146C
Inconclusive
Indiv-C
146C
Inconclusive

70

Table B4. mtDNA profiles obtained from spent cartridge casings loaded by volunteer OO
during Collection 1.
mtDNA
Extract
Haplotype
Assignment
OO
16093C, 16224C, 16311C, 73G
1A
152Y
Inconsistent
1A-MBI
16189Y, 16311Y, 73R, 152Y
Inconsistent
1B
16051G, 16129C, 16183C, 16189C, 73G, 152C
Inconclusive
1B-MBI
16051G, 16129S, 16183C, 16189C, 16278Y, 73G, 152C
Inconclusive
16126C, 16187T, 16189C, 16264T, 16270T, 16278T,
1C
16293G, 16311C, 73G, 152C, 185T, 189G, 195C, 247A
Inconclusive
16126C, 16187T, 16189C, 16264T, 16270T, 16278T,
1C-MBI
16293G, 16311C, 73G, 152C, 185T, 189G, 195C, 247A
Inconclusive
16093C, 16278T, 16294T, 16309G, 16311Y, 73G, 152C,
1D
195Y
Inconclusive
16093Y, 16278Y, 16294Y, 16309R, 16311Y, 16368Y, 73G,
1D-MBI
146Y, 152C, 195Y
Inconclusive
All-C
16189C, 16278T, 16311C, 73G, 152C
Inconclusive
Indiv-C
16189C, 16278T, 16311C, 73G, 152C
Inconclusive

Table B5. mtDNA profiles obtained from spent cartridge casings loaded by volunteer B during
Collection 2.
mtDNA
Extract
Haplotype
Assignment
B
16126C, 16294T, 16296T, 16304C, 16362C, 73G
24A
16126C, 16294T, 16296T, 16304C, 16362C, 73G
Consistent-Single
24B
16126C, 73R
Inconclusive
24B-MBI
16126C, 16294Y, 73R
Inconclusive
24C
16126C, 16294T, 16296T, 16304C, 16362C, 73G
Consistent-Single
24D
16126C, 16294T, 16296T, 16304C, 16362C, 73G, 228A
Inconclusive
16069Y, 16126C, 16294Y, 16296Y, 16304C, 16362Y, 73G,
24D-MBI
185R, 228R
Consistent-Multiple
All-C
16126C, 16294T, 16296T, 16304C, 16362C, 73G
Consistent-Single
Indiv-C
16126C, 16294T, 16296T, 16304C, 16362C, 73G
Consistent-Single

71

Table B6. mtDNA profiles obtained from spent cartridge casings loaded by volunteer C during
Collection 2.
mtDNA
Extract
Haplotype
Assignment
C
16274A, 152C
27A
16126Y, 16274A, 16294Y, 152C
Consistent-Single
27B
16126C, 16294T, 16304C, 73G, 183G
Inconsistent
27B-MBI
16126C, 16294T, 16304C, 73G, 183R
Inconsistent
27C
73R, 152Y
Inconsistent
27C-MBI
16274R, 16294Y, 73R, 152Y
Consistent-Multiple
27D
73R, 152Y
Inconsistent
27D-MBI
16291Y, 16294Y, 73R, 152Y
Inconsistent
All-C
16294T, 73G, 152C
Inconclusive
Indiv-C
16294T, 73G, 152Y
Inconclusive

Table B7. mtDNA profiles obtained from spent cartridge casings loaded by volunteer E during
Collection 2.
mtDNA
Extract
Haplotype
Assignment
E
16192T, 16256T, 16270T, 73G
8A
16192T, 73G
Inconclusive
8A-MBI
16192T, 16294Y, 73G
Inconclusive
8B
16192T, 16256T, 16270T, 73G
Consistent Single
8C
16126Y, 16163R, 16186Y, 16189Y, 16192Y, 73G
Inconclusive
8C-MBI
16126Y, 16163R, 16186Y, 16189Y, 16192Y, 73G
Inconclusive
8D
16192T, 16256T, 16270T, 73G
Consistent-Single
All-C
16192T, 73G
Inconclusive
Indiv-C
16192T, 16256T, 16270T, 73G
Consistent-Single

72

Table B8. mtDNA profiles obtained from spent cartridge casings loaded by volunteer H during
Collection 2.
mtDNA
Extract
Haplotype
Assignment
16145A, 16188T, 16189C, 16193T, 16193.1C, 16256T,
H
16270T, 16311C, 73G, 195C
16145A, 16188T, 16189C, 16193T, 16193.1C, 16256T,
32A
16270T, 16311C, 73G, 195C
Consistent-Single
16145A, 16188T, 16189C, 16193T, 16193.1C, 16256T,
32B
16270T, 16311C, 73G, 195C
Consistent-Single
16145R, 16188T, 16189C, 16193Y, 16193.1C, 16256Y,
32C
16270T, 16311C, 73G, 195Y
Consistent-Single
16145A, 16188T, 16189C, 16193T, 16193.1C, 16256T,
32D
16270T, 16311C, 73G, 195C
Consistent-Single
16145A, 16188T, 16189C, 16193T, 16193.1C, 16256T,
All-C
16270T, 16311C, 73G, 195C
Consistent-Single
16145A, 16188T, 16189C, 16193T, 16193.1C, 16256T,
Indiv-C
16270T, 16311C, 73G, 195C
Consistent-Single

Table B9. mtDNA profiles obtained from spent cartridge casings loaded by volunteer I during
Collection 2.
mtDNA
Extract
Haplotype
Assignment
I
16222T, 16235G, 16291T
11A
16222T, 16235G, 16291T
Consistent-Single
11B
16222T, 16235G, 16291T
Consistent-Single
11C
16222T, 16343G, 73G, 150T
Inconclusive
11C-MBI 16222Y, 16235R, 16291Y, 16343R, 73G, 150T, 156R, 160R
Inconclusive
11D
16222T, 16235G, 16291T
Consistent-Single
All-C
16222T, 16235G, 16291T
Consistent-Single
Indiv-C
16222T, 16235G, 16291T
Consistent-Single

73

Table B10. mtDNA profiles obtained from spent cartridge casings loaded by volunteer J during
Collection 2.
mtDNA
Extract
Haplotype
Assignment
J
16357C
21A
16357C
Consistent-Single
21B
16357C
Consistent-Single
21C
16357C
Consistent-Single
21D
16357C
Consistent-Single
All-C
16357C
Consistent-Single
Indiv-C
16357C
Consistent-Single

Table B11. mtDNA profiles obtained from spent cartridge casings loaded by volunteer L during
Collection 2.
mtDNA
Extract
Haplotype
Assignment
L
16189C, 16362C, 152C
22A
16179T, 16294T, 73G
Inconclusive
22A-MBI
16179Y, 16294T, 73G
Inconclusive
22B
16069Y, 16126C, 16362Y, 73G
Inconclusive
22B-MBI
16069Y, 16126Y, 16189Y, 16362Y, 73G, 185R, 228R
Inconsistent
22C
16069T, 16126C, 73G
Inconclusive
22C-MBI
16069T, 16126C, 16186Y, 73G
Inconclusive
22D
16126C, 73G
Inconclusive
22D-MBI
16069Y, 16126Y, 16189Y, 16362Y, 73G, 185R, 228R
Inconsistent
All-C
16069T, 16126C, 73G
Inconclusive
Indiv-C
16069T, 16126C, 16189Y, 16362Y, 73G, 185R, 228R
Inconsistent

74

Table B12. mtDNA profiles obtained from spent cartridge casings loaded by volunteer N during
Collection 2.
mtDNA
Extract
Haplotype
Assignment
N
Ø
25A
16235G, 16291T
Inconclusive
25A-MBI
16235G, 16291T, 73Y, 195Y
Inconclusive
25B
Ø
Consistent-Multiple
25C
Ø
Consistent-Multiple
25D
Ø
Consistent-Multiple
All-C
Ø
Consistent-Multiple
Indiv-C
Ø
Consistent-Multiple

Table B13. mtDNA profiles obtained from spent cartridge casings loaded by volunteer T during
Collection 2.
mtDNA
Extract
Haplotype
Assignment
T
16224C, 93G
15A
16186Y, 16189Y, 16224Y, 93R
Consistent-Multiple
15B
16294T, 73R
Inconclusive
15B-MBI
16294Y, 73R
Inconsistent
15C
Ø
Inconsistent
15C-MBI
Ø
Inconsistent
15D
16069T, 73G, 185R, 228R
Inconsistent
15D-MBI
16069Y, 16126Y, 16224Y, 73R, 93R, 185R, 228R
Consistent-Multiple
All-C
Ø
Inconsistent
Indiv-C
73R
Inconsistent

75

Table B14. mtDNA profiles obtained from spent cartridge casings loaded by volunteer BB
during Collection 2.
mtDNA
Extract
Haplotype
Assignment
BB
16298C, 16318C, 72C
31A
16298C, 16318M, 72C
Consistent-Single
31B
16189Y, 73Y, 151Y, 152Y
Inconsistent
16189Y, 16256Y, 16298Y, 16311Y, 16318M, 16362Y, 72Y,
31B-MBI
73R, 151Y, 152Y
Consistent-Multiple
31C
16298Y, 72C
Inconclusive
16189Y, 16256Y, 16270Y, 16298Y, 16311Y, 16318M,
31C-MBI
16362Y, 72Y, 73R
Consistent-Multiple
31D
16093Y, 16298Y, 195Y
Inconsistent
16093Y, 16256Y, 16270Y, 16298Y, 16311Y, 16318M,
31D-MBI
16362Y, 195C
Inconclusive
All-C
16298C, 16318M, 72C
Consistent-Single
16189Y, 16256Y, 16270Y, 16298Y, 16311Y, 16318M, 72Y,
Indiv-C
73R
Consistent-Multiple

Table B15. mtDNA profiles obtained from spent cartridge casings loaded by volunteer EE
during Collection 2.
mtDNA
Extract
Haplotype
Assignment
EE
Ø
18A
Ø
Consistent-Multiple
18B
73R
Consistent-Multiple
18C
73G
Inconclusive
18C-MBI
73G
Inconclusive
18D
Ø
Consistent-Multiple
All-C
Ø
Consistent-Multiple
Indiv-C
73G
Inconclusive

76

Table B16. mtDNA profiles obtained from spent cartridge casings loaded by volunteer FF
during Collection 2.
mtDNA
Extract
Haplotype
Assignment
FF
16093C, 16189C, 16270T, 73G, 150T
35A
16189Y, 16270T, 73G, 150Y
Inconclusive
35A-MBI
16093Y, 16189Y, 16270Y, 16291Y, 73R, 150Y
Consistent-Multiple
35B
16093Y, 16189Y, 16270T, 73G, 150T
Consistent-Single
35C
16235G, 16291T, 73G
Inconclusive
35C-MBI
16093Y, 16189Y, 16235R, 16270Y, 16291Y, 73G, 183R
Inconclusive
35D
16270Y, 150Y
Inconclusive
35D-MBI
16093Y, 16189Y, 16270Y, 16291Y, 73R, 150Y
Consistent-Multiple
All-C
16093Y, 16189Y, 16270T, 16291Y, 73G, 150T
Consistent-Single
Indiv-C
16093Y, 16189Y, 16270T, 16291Y, 73G, 150T
Consistent-Single

Table B17. mtDNA profiles obtained from spent cartridge casings loaded by volunteer GG
during Collection 2.
mtDNA
Extract
Haplotype
Assignment
GG
152C
16A
152C
Consistent-Single
16B
152C
Consistent-Single
16C
152C
Consistent-Single
16D
152C
Consistent-Single
All-C
152C
Consistent-Single
Indiv-C
152C
Consistent-Single

Table B18. mtDNA profiles obtained from spent cartridge casings loaded by volunteer II during
Collection 2.
mtDNA
Extract
Haplotype
Assignment
II
16278T, 16311C, 146C
34A
16278T, 16311C, 146C
Consistent-Single
34B
16278T, 16311C, 146C
Consistent-Single
34C
16278T, 16311C, 146C
Consistent-Single
34D
16278T, 16311C, 146Y
Consistent-Single
All-C
16278T, 16311C, 146C
Consistent-Single
Indiv-C
16278T, 16311C, 146C
Consistent-Single

77

Table B19. mtDNA profiles obtained from spent cartridge casings loaded by volunteer JJ during
Collection 2.
mtDNA
Extract
Haplotype
Assignment
JJ
Ø
9A
Ø
Consistent-Multiple
9B
Ø
Consistent-Multiple
9C
Ø
Consistent-Multiple
9D
Ø
Consistent-Multiple
All-C
Ø
Consistent-Multiple
Indiv-C
Ø
Consistent-Multiple

Table B20. mtDNA profiles obtained from spent cartridge casings loaded by volunteer NN
during Collection 2.
mtDNA
Extract
Haplotype
Assignment
16069T, 16126C, 16145A, 16172C, 16222T, 16261T, 73G,
NN
242T
16069T, 16126C, 16145A, 16172C, 16222T, 16261T, 73G,
37A
242T
Consistent-Single
16069T, 16126C, 16145A, 16172C, 16222T, 16261T, 73G,
37B
242T
Consistent-Single
16069T, 16126C, 16145A, 16172C, 16222T, 16261T, 73G,
37C
242T
Consistent-Single
16069T, 16126C, 16145A, 16172C, 16222T, 16261T, 73G,
37D
242T
Consistent-Single
16069T, 16126C, 16145A, 16172C, 16222T, 16261T, 73G,
All-C
242T
Consistent-Single
16069T, 16126C, 16145A, 16172C, 16222T, 16261T, 73G,
Indiv-C
242T
Consistent-Single

78

Table B21. mtDNA profiles obtained from spent cartridge casings loaded by volunteer PP
during Collection 2.
mtDNA
Extract
Haplotype
Assignment
PP
16311C, 152C
17A
16311C, 152C
Consistent-Single
17B
16311C, 152C
Consistent-Single
17C
16311C, 152C
Consistent-Single
17D
16311C, 152C
Consistent-Single
All-C
16311C, 152C
Consistent-Single
Indiv-C
16311C, 152C
Consistent-Single

Table B22. mtDNA profiles obtained from spent cartridge casings loaded by volunteer RR
during Collection 2.
mtDNA
Extract
Haplotype
Assignment
RR
16224C, 16311C, 73G, 199C
23A
16311C
Inconclusive
23A-MBI
16189Y, 16224*, 16261Y, 16311Y, 73R, 103R
Inconclusive
23B
16311C, 73R
Inconclusive
23B-MBI
16224Y, 16311Y, 73R, 199Y
Consistent-Multiple
23C
16224Y, 16311C, 73R
Inconclusive
23C-MBI
16224C, 16261Y, 16270Y, 16311C, 73R, 152Y
Inconclusive
23D
73G, 183G
Inconclusive
23D-MBI
16294Y, 16311Y, 73G, 183G
Inconclusive
All-C
16224N/A, 16311C, 73G
Inconclusive
Indiv-C
16224C, 16311C, 73G
Inconclusive
*base was not identifiable from the electropherograms

79

Table B23. mtDNA profiles obtained from spent cartridge casings loaded by volunteer SS
during Collection 2.
mtDNA
Extract
Haplotype
Assignment
SS
16126C, 16294T, 16304C, 73G, 183R
33A
16126C, 16294T, 16304C, 73G, 183G
Consistent-Single
33B
16126C, 16294T, 16304C, 73G
Consistent-Single
33C
16126C, 16294T, 16304C
Inconclusive
33C-MBI
16126C, 16294T, 16304C, 73R, 150Y
Consistent-Single
All-C
16126C, 16294T, 16304C, 73G
Inconsistent
Indiv-C
16126C, 16294T, 16304C, 73G
Inconsistent

Table B24. mtDNA profiles obtained from spent cartridge casings loaded by volunteer TT
during Collection 2.
mtDNA
Extract
Haplotype
Assignment
TT
16224C, 16261Y, 16270T, 16311C, 73G, 146C, 152C
19A
16224C, 16261T, 16270T, 16311C, 73G, 146C, 152C
Consistent-Single
19B
16224C, 16261T, 16270T, 16311C, 73G, 146C, 152C
Consistent-Single
19C
16224C, 16261Y, 16270T, 16311C, 73G, 146C, 152C
Consistent-Single
19D
16224C, 16270T, 16311C, 73G, 146C, 152C
Consistent-Single
All-C
16224C, 16261T, 16270T, 16311C, 73G, 146C, 152C
Consistent-Single
Indiv-C
16224C, 16261T, 16270T, 16311C, 73G, 146C, 152C
Consistent-Single

80

Table B25. mtDNA profiles obtained from spent cartridge casings loaded by volunteer UU
during Collection 2.
mtDNA
Extract
Haplotype
Assignment
UU
16129A, 16311C, 16316G
38A
16126C, 16294T, 16304C, 73G, 185R, 228R
Inconsistent
38A-MBI
16126Y, 16294Y, 16304Y, 73G, 183R, 185R, 228R
Inconsistent
38B
16126C, 16294Y, 73G
Inconclusive
38B-MBI
16126Y, 16294Y, 16304Y, 16311Y, 16316R, 73R, 150Y
Inconsistent
38C
16126C, 16189Y, 16294T, 73G
Inconclusive
38C-MBI
16126C, 16163R, 16186Y, 16189Y, 16294Y, 73R, 183R
Inconclusive
38D
16126C, 16294T, 73G
Inconclusive
38D-MBI
16126C, 16294Y, 16304Y, 73R, 183R
Inconsistent
All-C
16126C, 16294Y, 16304Y, 73G, 183R
Inconsistent
Indiv-C
16126C, 16294Y, 16304Y, 73R, 183R
Inconsistent

Table B26. mtDNA profiles obtained from spent cartridge casings loaded by volunteer VV
during Collection 2.
mtDNA
Extract
Haplotype
Assignment
VV
16069T, 16126C, 73G, 185A, 228A
20A
16069T, 16126C, 73G, 185A, 228A
Consistent-Multiple
20B
16069T, 16126C, 73G, 185A, 228A
Consistent-Multiple
20C
16069T, 16126C, 73G, 185R
Inconclusive
20C-MBI
16069T, 16126C, 16146R, 73G, 185R
Inconclusive
20D
16069T, 16126C, 73G, 185A, 228A
Consistent-Multiple
All-C
16069T, 16126C, 73G, 185A, 228A
Consistent-Multiple
Indiv-C
16069T, 16126C, 73G, 185A, 228A
Consistent-Multiple

81

Table B27. mtDNA profiles obtained from spent cartridge casings loaded by volunteer WW
during Collection 2.
mtDNA
Extract
Haplotype
Assignment
16189C, 16192T, 16256T, 16270T, 16311C, 16362C, 73G,
WW
151T, 152C
30A
16256Y, 16270T, 16311Y, 16362C
Inconclusive
30A-MBI
16256Y, 16270T, 16311Y, 16362C, 73Y, 151Y, 152Y
Inconsistent
16189C, 16192Y, 16256T, 16270T, 16311C, 16362C, 73G,
30B
151T, 152C
Consistent-Single
16189C, 16192T, 16256T, 16270T, 16311C, 16362C, 73G,
30C
151T, 152C
Consistent-Single
16126Y, 16189C, 16192Y, 16256T, 16270T, 16311C,
30D
16363C, 73G, 151T, 152C
Consistent-Single
16189C, 16192T, 16256T, 16270T, 16311C, 16362C, 73G,
All-C
151T, 152C
Consistent-Single
16189C, 16192T, 16256T, 16270T, 16311C, 16362C, 73G,
Indiv-C
151T, 152C
Consistent-Single

Table B28. mtDNA profiles obtained from spent cartridge casings loaded by volunteer XX
during Collection 2.
mtDNA
Extract
Haplotype
Assignment
XX
16069T, 16126Y, 73G, 185A, 228A
36A
16069T, 16126C, 73G, 185A, 228A
Consistent-Multiple
36B
16069T, 16126C, 73G, 185A, 228A
Consistent-Multiple
36C
16069T, 16126C, 73G, 185A, 228A
Consistent-Multiple
36D
16126C, 16163R, 16189Y, 73G, 152Y, 185R, 228A
Inconclusive
16126Y, 16163R, 16186Y, 16189Y, 73R, 152Y, 178R, 185R,
36D-MBI
228R
Inconsistent
All-C
16069T, 16126C, 73G, 185A, 228A
Consistent-Multiple
Indiv-C
16069T, 16126C, 73G, 185A, 228A
Consistent-Multiple

82

REFERENCES

83

REFERENCES

Abaz J, Walsh SJ, Curran JM, Moss DS, Cullen J, Bright JA et al. Comparison of the variables
affecting the recovery of DNA from common drinking containers. Forensic Sci Int
2002;126:233 – 40.
Addinsoft. XLSTAT 1995 – 2013.
Alessandrini F, Cecati M, Pesaresi M, Turchi C, Carle F, Tagliabracci A. Fingerprints as
evidence for a genetic profile: morphological study on fingerprints and analysis of
exogenous and individual factors affecting DNA typing. J Forensic Sci 2003;48(3):586
92.
Ammunition Basics. http://www.gunclassics.com/ammunition.html.
Anderson S, Bankier AT, Barrell BG, de Bruijn MH, Coulson AR, Drouin J et al. Sequence and
organization of the human mitochondrial genome. Nature 1981;290(5806):457 – 65.
Applied Biosystems. AmpFℓSTR®Identifiler® PCR amplification kit, Applied Biosystems
2012a Foster City, CA.
Applied Biosystems. AmpFℓSTR®Identifiler® Plus PCR amplification kit, Applied
Biosystems 2012b Foster City, CA.
Applied Biosystems. AmpFℓSTR® MiniFiler™ PCR amplification kit user’s manual, Applied
Biosystems 2012c Foster City, CA.
Bender K, Farfan MJ, Schneider PM. Preparation of degraded human DNA under controlled
conditions. Forensic Sci Int 2004;139:135 – 40.
Bentsen RK, Brown JK, Dinsmore A, Harvey KK, Kee TG. Post firing visualization of
fingerprints on spent cartridge cases. Sci Justice 1996;36(1):3 – 8.
Bingham A. Shootings that shaped gun control laws. Abcnews.go.com 27 July 2012.
Bodenteich A, Mitchell LG, Merril CR. A lifetime of retinal light exposure does not appear to
increase mitochondrial mutations. Gene 1991;108(2):305 – 9.
Branch L. Generating forensic DNA profiles from ‘contact’ DNA on cartridge cases and gun
grips [thesis]. Sacramento (CA): CA State Univ., Sacramento, 2010.
Brevnov MG, Pawar HS, Mundt J, Calandro LM, Furtado MR, Shewale JG. Developmental
validation of the PrepFiler™ forensic DNA extraction kit for extraction of genomic
DNA from biological samples. J Forensic Sci 2009;54(3):599 – 607.
84

Cork DL, Nair VN, Rolph JE. Some forensic aspects of ballistic imaging. Fordham Urb LJ
2010;38(2):473 – 501.
Daly DJ, Murphy C, McDermott SD. The transfer of touch DNA from hands to glass, fabric and
wood. Forensic Sci Int: Gen 2012;6:41 – 6.
Dieltjes P, Mieremet R, Zuniga S, Kraaijenbrink T, Pijpe J, de Knijff P. A sensitive method to
extract DNA from biological traces present on ammunition for the purpose of genetic
profiling. Int J Legal Med 2011;125:597 – 602.
Diffinity Genomics. Diffinity RapidTip user handbook. Diffinity Genomics 2012 West
Henrietta, NY.
Dufour C and Dangles O. Flavonoid-serum albumin complexation: determination of binding
constants and binding sites by fluorescence spectroscopy. Biochimica et Biophysica
Acta 2005;1721:164 – 73.
Dulan S. Piers Morgan Tonight. 14 Dec 2012.
Edson SM, Ross JP, Coble MD, Parsons TJ, and Barritt SM. Naming the dead—confronting the
realities of rapid identification of degraded skeletal remains. Forensic Sci Rev
2004;16:63 – 90.
Eilert KD, Foran DR. Polymerase resistance to polymerase chain reaction inhibitors in bone. J
Forensic Sci 2009;54(5):1001 – 7.
Elbe RE. External barrel temperature of the M16A1 rifle. Small arms weapons systems
directorate, technical report 1975, Rock Island, IL.
Esslinger KJ, Siegel JA, Spillane H, Stallworth S. Using STR analysis to detect human DNA
from exploded pipe bomb devices. J Forensic Sci 2004;49(3):481 – 4.
Fasano M, Curry S, Terreno E, Galliano M, Fanali G, Narciso P et al. The extraordinary ligand
binding properties of human serum albumin. Life 2005;57(12):787 – 96.
FBI. Uniform Crime Reports. 2012.
http://www.fbi.gov/about-us/cjis/ucr/crime-in-theu.s/2011/crime-in-the-u.s.-2011.
FBI. Using Science to Solve Crimes. 2006.
www.fbi.gov/about-us/lab/codis/codis-and-ndisfact-sheet.
Fisher Scientific. ELIMINase® Decontaminant technical data sheet. Decon Laboratories, Inc.,
Bryn Mawr, PA.
Foran DR. Relative degradation of nuclear and mitochondrial DNA: an experimental
approach. J Forensic Sci 2006;51(4):766 – 70.

85

Freund H. New thug fad: gun share-and-stash. Nypost.com 22 June 2012.
Frieden T. US violent crime down for fifth straight year. CNN Justice 29 Oct 2012.
Gabriel MN, Huffine EF, Ryan JH, Holland MM, Parsons TJ. Improved mtDNA sequence
analysis of forensic remains using a “mini-primer set” amplification strategy. J Forensic
Sci 2001;46(2):247 – 53.
Gashi B, Edwards MR, Sermon PA, Courtney L, Harrison D, Xu Y. Measurement of 9mm
cartridge case external temperatures and its forensic application. Forensic Sci Int
2010;200(1 – 3):21 – 7.
Gill P. Application of low copy number DNA profiling. Croatian Med J 2001;42(3):229 – 32.
Given BW. Latent fingerprints on cartridges and expended cartridge casings. J Forensic Sci
1976;21(3):587 – 94.
Goforth W. History of gun control in the US. Krctv.com 19 Feb 2013.
Goodwin W, Linacre A, Vanezis P. The use of mitochondrial DNA and short tandem repeat
typing in the identification of air crash victims. Electrophoresis 1999;20:1707 – 11.
Goray M, Eken E, Mitchell RJ, van Oorschot RAH. Secondary DNA transfer of biological
substances under varying test conditions. Forensic Sci Int: Gen 2010;4:62 – 7.
Hall TA. BioEdit: a user-friendly biological sequence alignment editor and analysis program for
Windows 95/98/NT. Nucleic Acids Symp Ser 1999;41:95 – 8.
Hebda LM, Doran AE, Foran DR. Collecting and analyzing DNA evidence from fingernails: a
comparative study. J Forensic Sci. In press.
Hoffmann SG, Stallworth SE, Foran DR. Investigative studies into the recovery of DNA from
improvised explosive device containers. J Forensic Sci 2012;57(3):602 – 9.
Holland MM, Parsons TJ. Mitochondrial DNA sequence analysis—validation and use for
forensic casework. Forensic Sci Rev 1999;11(1):22 – 50.
Horsman-Hall KM, Orihuela Y, Karczynski SL, Davis AL, Ban JD, and Greenspoon SA.
Development of STR profiles from firearms and fired cartridge cases. Forensic Sci Int:
Gene 2009;3:242 – 50.
Hoyert DL, Xu J. Death: preliminary data for 2011. National Vital Statistics Reports (CDC)
2012;61(6):19.
Karni M, Zidon D, Polak P, Zalevsky Z, Shefi O. Thermal degradation of DNA. DNA Cell Biol
2013;32(6):298 –301.

86

Kita T, Yamaguchi H, Yokoyama M, Tanaka T, Tanaka N. Morphological study of fragmented
DNA on touched objects. Forensic Sci Int: Gen 2008;3:32 – 6.
Krampen AA, Ells DO, Bjerke RK, and Ward JP, inventors. Omark Industries, Inc., assignee.
Primer composition. US patent 4,608,102. 1986 Aug 26.
Kreader CA. Relief of amplification inhibition in PCR with bovine serum albumin or T4 gene
32 protein. Appl Environ Microbiol 1996;62(3):1102 – 6.
Krouse WJ. Gun Control Legislation. Congressional Research Service. 2012.
Ladd C, Adamowicz MS, Bourke MT, Scherczinger CA, Lee HC. A systematic analysis of
secondary DNA transfer. J Forensic Sci 1999;44(6):1270 – 2.
Lee HY, Kim NY, Park MJ, Yang WI, Shin KJ. A modified mini-primer set for analyzing
mitochondrial DNA control region sequences from highly degraded forensic samples.
Biotechniques 2008;44:555 – 8.
Locard E. Dust and its analysis. Police J 1928;1(2):177 – 92. Referenced by Saferstein R.
Forensic Science Handbook, vol. II. Englewood Cliffs, NJ. 1988.
Lowe A, Murray C, Whitaker J, Tully G, Gill P. The propensity of individuals to deposit DNA
and secondary transfer of low level DNA from individuals to inert surfaces. Forensic Sci
Int 2002;129:25 – 34.
Matte M, Williams L, Frappier R, Newman J. Prevalence and persistence of foreign DNA
beneath fingernails. Forensic Sci Int: Gen 2012;6:236 – 43.
Melton T, Holland C, Holland M. Forensic mitochondrial DNA analysis: current practice and
future potential. Forensic Sci Rev 2012;24:101 – 22.
Milloy S. How reliable is ballistic fingerprinting?. Foxnews.com 18 Oct 2002.
Monnat RJ, Loeb LA. Nucleotide sequence preservation of human mitochondrial DNA. Proc
Natl Acad Sci USA 1985;82(9):2895 – 9.
Monnat RJ, Reay DT. Nucleotide sequence identity of mitochondrial DNA from different
human tissues. Gene 1986;43(3):205 – 11.
National Forensic Science Technology Center.
http://www.nfstc.org/pdi/Subject09/pdi_s09_m01_01_b.htm.
Nickerson DA, Tobe VO, Taylor SL. PolyPhred: automating the detection and genotyping of
single nucleotide substitutions using fluorescence-based resequencing. Nucleic Acids
Res 1997;25(14):2745 – 51.

87

Nunn S. Touch DNA collection versus firearm fingerprinting: comparing evidence production
and identification outcomes. J Forensic Sci 2013;58(3):601 – 8.
Orlando A. The recovery and analysis of DNA from fired cartridge casings [thesis]. East
Lansing (MI): Michigan State Univ. 2012.
Pizzamiglio M, Mameli A, My D, Garofano L. Forensic identification of a murderer by LCN
DNA collected from the inside of the victim’s car. Int Congr Ser 2004;1261:437 – 9.
Promega. PowerPlex® 16 System technical manual, Promega 2013 Madison, WI.
Raymond JJ, van Oorschot RAH, Walsh SJ, Roux C, Gunn PR. Trace DNA and street robbery:
a criminalistics approach to DNA evidence. Forensic Sci Int: Gen Suppl Ser
2009;2(1):544 – 6.
Richardson DA, Kosa R. An examination of homicide clearance rates: foundation for the
development of a homicide clearance model. Washington, DC; Police Executive
Research Forum. 2001.
Richert NJ. Swabbing firearms for handler’s DNA. J Forensic Sci 2011;56(4):972 – 5.
R Core Team. R: a language and environment for statistical computing. Vienna, Austria 2013.
www.r-project.org.
Roberts D, Marks R. The determination of regional and age variations in the rate of
desquamation: a comparison of four techniques. J Invest Dermatol 1980;74:13 – 6.
Robin ED, Wong R. Mitochondrial DNA molecules and virtual number of mitochondria per cell
in mammalian cells. J Cell Physiol 1988;136(3):507 – 13.
Satoh M, Kuroiwa T. Organization of multiple nucleoids and DNA molecules in mitochondria
of a human cell. Exp Cell Res 1991;196(1):137 – 40.
Schulz MM, Reichert W. Archived or directly swabbed latent fingerprints as a DNA source for
STR typing. Forensic Sci Int 2002;127:128 – 30.
Sigursardottir S, Helgason A, Gulcher JR, Stefansson K, Donnelly P. The mutation rate in the
human mtDNA control region. Am J Hum Genet 2000;66:1599 – 609.
Sparano VT. The complete outdoors encyclopedia. 4th ed. New York: St. Martin’s Press, 2000.
Spear T, Clark J, Giusto M, Khoshkebari N, Murphy M, and Rush J. Fingerprints and cartridge
cases: how often are fingerprints found on handled cartridge cases and can these
fingerprints be successfully typed for DNA?. Proceedings of California Association of
Criminalistics; 2005 May 9 – 15; Oakland, CA. Sacramento: California Criminalistics
Institute, 2005.

88

Sweet D, Lorente M, Lorente JA, Valenzuela A, Villanueva E. An improved method to recover
saliva from human skin: the double swab technique. J Forensic Sci 1997;42(2):320 – 2.
Thomasma SM, Foran DR. The influence of swabbing solutions on DNA recovery from touch
samples. J Forensic Sci 2013;58(2):465 – 9.
US Department of Justice. mtDNA population database user’s manual. Forensic Sci
Communication 2002;4(2).
van Oorschot RAH, Jones MK. DNA fingerprints from fingerprints. Nature 1997;387:767.
Viray JL. Implementation of Applied Biosystems AmpFlSTR® MiniFiler™ amplification kit
for forensic casework: validation, inhibition, and contact DNA studies [thesis]. Davis
(CA): Univ. of CA, Davis, 1998.
Westrom MA, Reynolds GL, Reynolds SP, inventors. ArmaLite, Inc., assignee. Cartridge for a
firearm. US patent 6,976,431. 2005 Dec 20.
Wilson MR, Polanskey D, Replogle J, DiZinno JA, Budowle B. A family exhibiting
heteroplasmy in the human mitochondrial DNA control region reveals both somatic
mosaicism and pronounced segregation of mitotypes. Hum Genet 1997;100:167 –71.

89