Genomics for Bioforensics
Abstract
The goal of the Genomics for Bioforensics project (FY06-FY08) has been to explore the application of genomics to the challenge of attribution for biological organisms: given a sample of an agent, can we use genomics to (help) determine whether the new sample is an endemic strain, a strain introduced from another place or time, or a novel (possibly engineered) strain? Specifically, given sequence data from a new sample, we have developed procedures to compare this "unknown" sequence to a reference database of sequences from previously collected samples and their associated metadata. To do this, we have created a reference database for a specific organism (influenza) and developed procedures to create a Microbial Forensics Workbench, which enables the user to compare the "unknown" strain with the strains in the Reference Database. The Workbench provides a novel automated method for clustering the "unknown" strain with the most similar strains in the Reference Database. This method for genotyping uses Complete Composition Vectors and Affinity Propagation Clustering (submitted for publication: Peterson et al., Bioinformatics). The Workbench also supports several visualization techniques, including display of a color-coded phylogenetic tree (using TreeViewJ, Colosimo et al., BMC Bioinformatics, also developed under this project), as well as map and timeline displays based on the Simile open source software.
Document Details
- Document Type
- Technical Report
- Publication Date
- Dec 01, 2008
- Accession Number
- AD1107406
Entities
People
- Joanne Luciano
- Lynette Hirschman
- Marc Colosimo
- Matthew Peterson
- Meredith Keybl
- Scott Mardis
Organizations
- MITRE Corporation