Taxonomic identification of biological materials can be achieved through DNA barcoding, where an unknown “barcode” sequence is compared to a reference database. In many disciplines, obtaining accurate taxonomic identifications can be imperative ( e . g ., evolutionary biology, food regulatory compliance, forensics). The Barcode of Life DataSystems (BOLD) and GenBank are the main public repositories of DNA barcode sequences. In this study, an assessment of the accuracy and reliability of sequences in these databases was performed. To achieve this, 1) curated reference materials for plants, macro-fungi and insects were obtained from national collections, 2) relevant barcode sequences ( rbcL , matK , trnH-psbA , ITS and COI ) from these reference samples were generated and used for searching against both databases, and 3) optimal search parameters were determined that ensure the best match to the known species in either database. While GenBank outperformed BOLD for species-level identification of insect taxa (53% and 35%, respectively), both databases performed comparably for plants and macro-fungi (~81% and ~57%, respectively). Results illustrated that using a multi-locus barcode approach increased identification success. This study outlines the utility of the BLAST search tool in GenBank and the BOLD identification engine for taxonomic identifications and identifies some precautions needed when using public sequence repositories in applied scientific disciplines.

Soil DNA profiling has potential as a forensic tool to establish a link between soil collected at a crime scene and soil recovered from a suspect. However, a quantitative measure is needed to investigate the spatial/temporal variability across multiple scales prior to their application in forensic science. In this study, soil DNA profiles across Miami-Dade, FL, were generated using length heterogeneity PCR to target four taxa. The objectives of this study were to (i) assess the biogeographical patterns of soils to determine whether soil biota is spatially correlated with geographic location and (ii) evaluate five machine learning algorithms for their predictive ability to recognize biotic patterns which could accurately classify soils at different spatial scales regardless of seasonal collection. Results demonstrate that soil communities have unique patterns and are spatially autocorrelated. Bioinformatic algorithms could accurately classify soils across all scales with Random Forest significantly outperforming all other algorithms regardless of spatial level.

Ecological studies of microbial communities often use profiling methods but the true community diversity can be underestimated in methods that separate amplicons based on sequence length using performance optimized polymer 4. Taxonomically, unrelated organisms can produce the same length amplicon even though the amplicons have different sequences. F-108 polymer has previously been shown to resolve same length amplicons by sequence polymorphisms. In this study, we showed F-108 polymer, using the ABI Prism 310 Genetic Analyzer and CE, resolved four bacteria that produced the same length amplicon for the 16S rRNA domain V3 but have variable nucleotide content. Second, a microbial mat community profile was resolved and supported by NextGen sequencing where the number of peaks in the F-108 profile was in concordance with the confirmed species numbers in the mat. Third, equine DNA was analyzed for SNPs. The F-108 polymer was able to distinguish heterozygous and homozygous individuals for the melanocortin 1 receptor coat color gene. The method proved to be rapid, inexpensive, reproducible, and uses common CE instruments. The potential for F-108 to resolve DNA mixtures or SNPs can be applied to various sample types-from SNPs to forensic mixtures to ecological communities.

Molecular methods for the detection of mammalian coat color phenotypes have expanded greatly within the past decade. Many phenotypes are associated with a single nucleotide polymorphism mutation in the genetic sequence. Traditionally, these mutations are detected through sequencing, hybridization assays or mini-sequencing. However, these techniques can be expensive and tedious. Previously, CE-SSCP using the F-108 polymer was able to distinguish SNPs for the melanocortin-1 receptor (mc1r) coat color gene in horses (Equus caballus) that differed by one nucleotide substitution. The objective of this study was to expand the detection of coat color SNPs in horses. The genes for the solute carrier family member 2 (slc45a2/matp), type III receptor protein-tyrosine kinase (kit) and mc1r genes using CE-SSCP and F-108 polymer were compared to mini-sequencing with the SNaPshot kit. The F-108 polymer reproducibly resolved homozygous and heterozygous individuals for the mc1r and kit markers, but was unable to resolve heterozygous individuals for slc45a2 at 38ºC. The need for temperatures <15ºC, the SNP position being close to the 5'-end, and conformational structures/free energy with similar values resulted in the inability to resolve the secondary structures. Despite this limitation, the CE-SSCP method could be used to provide a rapid phenotypic description for equine forensic investigations.

Detection of seed-based toxins is a need for forensic chemists when suspected poisonings occur. The evidence that is found is often physically unidentifiable, as the seeds are mashed to extract the toxins. This work investigates potential strategies for rapid detection of seed-based toxins using chemical signatures obtained by direct analysis in real time mass spectrometry (DART-MS). Seven toxins (digoxin, digitoxin, hypaconitine, hyoscyamine, lanatoside, oleandrin, and scopolamine) and six seeds containing these toxins were studied. While detection of four of the toxins was readily attainable, detection of digoxin, digitoxin, and lanatoside was hindered by the inability to thermally desorb these larger compounds under normal operating conditions. The use of DART-MS variants capable of higher temperature analysis (thermal desorption (TD)-DART-MS and infrared thermal desorption ((IRTD)-DART-MS) enabled detection of these compounds. The low-level toxin concentrations and limited number of seeds available for analysis led to detection difficulties from both seed mashes and methanolic seed mash extracts. Principal component analysis (PCA) of generated mass spectra enabled differentiation of seed species, even in the cases when the toxins were undetectable.