Analytical and bioanalytical chemistry measurement results are most meaningful when interpreted using rigorous statistical treatments of the data. The same data set may provide many dimensions of information depending on the questions asked through the applied statistical methods. Three principal projects illustrated the wealth of information gained through the application of statistical data analyses to diverse problems. Firstly, novel aerosol test particles containing DNA barcodes were developed for the accurate assessment of aerosol transport and fate in populated locations. Aerosols are central to human and environmental health, and understanding the properties of these aerosols that are pervasive in our lives is essential. Test particles composed of FDA-approved saccharide food additives were generated using both a modified inkjet printer and a commercial spray dryer. Univariate statistical methods were used to evaluate generated particle size-distributions during production optimization. Non-coding DNA templates were incorporated into the particles as unique particle identifiers, which yielded customized test particles detectable using highly specific quantitative real-time polymerase chain reaction (QRT-PCR) assays. These safe, customizable, and specifically detected aerosol test particles will provide vital experimental feedback for evaluating aerosol dispersion and transport models. The project culminated with a successful demonstration of the aerosol test particles in an atmospheric release test. Secondly, an original method for non-invasively analyzing the chemical profiles of latent fingerprint residues was developed in order to gain a new level of information from the most common type of forensic evidence. Passive solid-phase microextraction (SPME) headspace sampling collects both endogenous and exogenous volatile and semi-volatile compounds contained in the fingerprint residue while preserving the fingerprint for traditional analyses. The information-rich chemical profiles obtained from gas chromatography-mass spectrometry (GC-MS) analyses of the SPME samples were used to quantitatively compare fingerprint compounds both between subjects and over a time course of 30 days using multivariate statistical analyses. Finally, endogenous metabolite profiles of cancer cells treated with anti-cancer agents were analyzed using gas chromatography- and high performance liquid chromatography- mass spectrometry (GC-MS and HPLC-MS), and the resulting complex data sets were interrogated using univariate and multivariate statistical analyses (e.g. ANOVA, PCA, PLS-DA, OPLS-DA). Possible modes of cytotoxicity of cisplatin and taxol, two commonly used cancer therapeutics, in breast and lung cancer cells were elucidated using statistical methods for data reduction in order to focus on the cellular biochemical processes most affected by drug treatment. Understanding how successful therapeutics interact with cells leads to design of novel anti-cancer agents that are more targeted and effective, minimizing dose-limiting side-effects and saving more lives. Although the areas of study are diverse, the commonality is the generation of highly complex data tables that may be effectively analyzed and interpreted using statistical methods.
Read
