USAFA Longitudinal Stability and Microbial Forensics. USAFA - Random forest modeling for microbial forensic applications utilizing longitudinal human and built environment microbiome inputs

The purpose of this study was to examine the forensic utility of the associations between the human microbiome and the microbiome of the built environment. Samples were collected from the hand and surfaces in office (n = 22) and home environments (n = 11) once a week for three consecutive weeks. This study aimed to: 1) determine the longitudinal stability of the skin and surface microbiomes over the three-week sampling periods; 2) determine if random forest models were accurate in predicting the occupant of an office or home; 3) determine if the accuracy of random forest models varied based on sample type and participant; 4) determine the accuracy of random forest models to predict an occupant across the office and home environments.