U.S. flag

An official website of the United States government, Department of Justice.

Improving Estimates of the Postmortem Interval with Metagenomics and Metabolomics

NCJ Number
300821
Date Published
2020
Length
18 pages
Author(s)
Jessica Lynne Metcalf; Rob Knight; Pieter Dorrestein; David O. Carter
Agencies
NIJ-Sponsored
Grant Number(s)
2016-DN-BX-4194
Annotation

The goal of this research project was to develop a three-pronged approach to investigating postmortem microbial ecology and improve understanding of the accuracy, reliability, and measurement validity of postmortem microbiome data.

Abstract

The use of molecular data has transformed forensic science by enabling the identification of individuals, unknown materials, and metabolic processes. These molecular approaches have had a significant impact on medicolegal death investigation because they have led to the identification of people who have died and the factors that contributed to and caused death. The power of these data can be further increased by incorporating metagenomic and metabolomics data. Comparing the three datasets (amplicon, shotgun metagenomics, and metabolites) will enable determination of 1) what microbes are present, 2) what the microbes are doing, and 3) what chemical compounds they are generating, respectively. The current project incorporated shotgun metagenomic data and metabolomics data into the workflow of a project that used thousands of samples in investigating the decomposition of human corpses in three contrasting environments in Colorado, Tennessee, and Texas. The project’s design will facilitate the achievement of both fundamental and applied research goals. The proposed datasets will be applied to a forensic context and represent one of the first studies to examine these components in disparate terrestrial ecosystems. To date, the project has identified several common human-associated metabolites and confirmed their presence across seasons. Analyses are underway to analyze additional human-associated metabolites and their persistence during decomposition. Although these data are still small for machine learning applications, as more data become available, these models will become more robust. 4 figures, 2 tables, and 20 references

Date Created: May 6, 2021