Forensic Investigative Genetic Genealogy (FIGG) has rapidly emerged as a powerful investigative tool. Since the arrest of the Golden State Killer in 2018, numerous forensic and law enforcement (LE) agencies have assessed, or applied, FIGG in unidentified human remains and criminal cases internationally. However, in its current form, FIGG can be a time consuming and resource intensive process which may place it outside the reach of some law enforcement agencies.
The application of FIGG, particularly in unidentified human remains cases, is as ongoing challenge. Generating whole genome single nucleotide polymorphism (SNP) data from compromised or degraded biological evidence often results in unusable SNP data not compatible for upload into LE permitted databases (GEDmatch Pro and FTDNA). An associated challenge is the complexity of building pedigree trees. Data from the US indicates that the average time for a FIGG case from initiation to resolution is approximately 12 months [1].This presents considerable resourcing issues for already stretched LE agencies. However, automation of the review, and triage, of match lists and family tree building has significant potential to expedite the initial phase of FIGG casework.
Forensic applications of artificial intelligence (AI) and machine learning (ML) are being explored more widely including in FIGG. A suite of tools which utilise AI and ML, developed by Indago Solutions (US), were assessed by the Australian Forensic Genetic Genealogy Collaboration (AFGGC) to determine their potential use in FIGG for coronial and criminal investigations. PARSNiP is a Java application that curates SNP data from commonly used direct-to-consumer service providers into formats more amenable to upload. PARSNiP also allows a higher level of user input to i) create a ‘super set’ of SNPs by merging WGA data sets and ii) develop a custom SNP set containing specific SNPs of interest for an individual. SNPCheck , the companion tool, specifically examines two or three SNP files and any identifies discrepancies. Indago’s toolkit also provides integration with open-source data to facilitate automated triage and assessment of match lists. Autogenerated family trees demonstrate estimated links between matches based on open-source information.
To assess the applicability of FIGG tools developed by Indago for international cases, the GEDmatch match lists of 14 known consenting Australian donors (research kits) were analysed. The results from this preliminary study suggests that AI and ML FIGG tools have the potential to significantly reduce the resource-intensive aspects of this process allowing valuable resources to be redirected to other areas of the investigative process to expedite FIGG-generated intelligence leads.
[1] Dowdeswell TL. Forensic genetic genealogy: A profile of cases solved. Forensic Sci Int Genet. 2022 May;58:102679.
Forensic Investigative Genetic Genealogy (FIGG) has rapidly emerged as a powerful investigative tool. Since the arrest of the Golden State Killer in 2018, numerous forensic and law enforcement (LE) agencies have assessed, or applied, FIGG in unidentified human remains and criminal cases internationally. However, in its current form, FIGG can be a time consuming and resource intensive process which may place it outside the reach of some law enforcement agencies.
The application of FIGG, particularly in unidentified human remains cases, is as ongoing challenge. Generating whole genome single nucleotide polymorphism (SNP) data from compromised or degraded biological evidence often results in unusable SNP data not compatible for upload into LE permitted databases (GEDmatch Pro and FTDNA). An associated challenge is the complexity of building pedigree trees. Data from the US indicates that the average time for a FIGG case from initiation to resolution is approximately 12 months [1].This presents considerable resourcing issues for already stretched LE agencies. However, automation of the review, and triage, of match lists and family tree building has significant potential to expedite the initial phase of FIGG casework.
Forensic applications of artificial intelligence (AI) and machine learning (ML) are being explored more widely including in FIGG. A suite of tools which utilise AI and ML, developed by Indago Solutions (US), were assessed by the Australian Forensic Genetic Genealogy Collaboration (AFGGC) to determine their potential use in FIGG for coronial and criminal investigations. PARSNiP is a Java application that curates SNP data from commonly used direct-to-consumer service providers into formats more amenable to upload. PARSNiP also allows a higher level of user input to i) create a ‘super set’ of SNPs by merging WGA data sets and ii) develop a custom SNP set containing specific SNPs of interest for an individual. SNPCheck , the companion tool, specifically examines two or three SNP files and any identifies discrepancies. Indago’s toolkit also provides integration with open-source data to facilitate automated triage and assessment of match lists. Autogenerated family trees demonstrate estimated links between matches based on open-source information.
To assess the applicability of FIGG tools developed by Indago for international cases, the GEDmatch match lists of 14 known consenting Australian donors (research kits) were analysed. The results from this preliminary study suggests that AI and ML FIGG tools have the potential to significantly reduce the resource-intensive aspects of this process allowing valuable resources to be redirected to other areas of the investigative process to expedite FIGG-generated intelligence leads.
[1] Dowdeswell TL. Forensic genetic genealogy: A profile of cases solved. Forensic Sci Int Genet. 2022 May;58:102679.
Workshop currently at capacity. A waitlist is available to join on our registration page.
CTO, Indago Solutions, LLC
Isak Bosman's pioneering work in AI has been recognized on a global stage, notably at the United Nations Convention in Geneva for combating child sex abuse. He has also been honored with an Innovation Award from the Air Force for his groundbreaking AI contributions to field operations. As CTO of Indago Solutions, Isak is committed to leveraging AI for good by developing solutions that aid investigators in solving critical crimes.
Co-founder, Indago Solutions, LLC.
Mr. Busch and his partner Steve Kramer left the FBI and founded Indago Solutions, an AI-based software as a service that automates genetic genealogy to produce investigative leads and identify suspects. He continues to work regularly with the US Department of Justice and with state and local investigators to solve cases through the use of genetic genealogy.
"*" indicates required fields
Copyright © 2024 ISHI. All Rights Reserved.
