Current forensic comparisons of soil most often rely upon physical characterizations. We hypothesized that bacterial community profiles obtained by Amplicon Length Heterogeneity-Polymerase Chain Reaction (ALH-PCR) of the 16S rRNA genes would provide discriminating data for soil comparisons. Dual extractions and replicate amplifications were performed on each soil. Chemical characterization by elemental analysis, pH, moisture content, percent Carbon and percent Nitrogen were performed. Supervised classification of the microbial community profiles using a Support Vector Machine (SVM) learning tool was over 95 % accurate labeling a microbial community profile to its originating soil type. By comparison, the chemical analysis data yielded accuracies between 40 and 77 %. The results of this study support the application of this method in the comparison of casework size soil samples. Results of this study may also justify the future development of a database of microbial community profiles for inferring the possible origin of unknown soil samples.
Identifer | oai:union.ndltd.org:fiu.edu/oai:digitalcommons.fiu.edu:etd-3874 |
Date | 03 December 2007 |
Creators | Crandall, Todd Martin |
Publisher | FIU Digital Commons |
Source Sets | Florida International University |
Detected Language | English |
Type | text |
Format | application/pdf |
Source | FIU Electronic Theses and Dissertations |
Page generated in 0.0016 seconds