Expert Systems: FSS-i3 Software Evaluation

Participants: Carolyn R. "Becky" Hill, Angela Dolph (Marshall University), Amy E. Decker, Margaret C. Kline, David L. Duewer, and John M. Butler

Project Timeframe: January 2006 to present

Purpose: To examine genotyping accuracy of expert system software and to develop tools that will aid use.

Progress: Forensic DNA testing typically involves two different types of samples: (1) single source reference samples from convicted offenders, suspected perpetrators or biological relatives of missing persons and (2) casework evidence that is often a mixture of victim and perpetrator and may be compromised in terms of quality and quantity of material.  While casework samples often are challenging to analyze, the shear number of the relatively easily analyzed reference samples is itself a bottleneck that requires development and validation of high-throughput data review procedures.  As of April 2007, the National DNA Index System maintained by the FBI Laboratory contained over 4.4 million convicted offender profiles but only 173,000 forensic casework profiles.  Thus, 96% of samples on the national DNA database are single source reference samples. In January 2006, the NIST Human Identity Project team purchased the Forensic Science Service FSS-i3 Expert System Software from Promega Corporation (Madison, WI).  Short tandem repeat (STR) typing data for over 1,000 samples have been evaluated thus far with the FSS-i3 software and compared with manually evaluated results.  Several Excel-based software tools have been developed to aid conversion of data formats and comparison of manually produced and expert system derived allele calls.  These tools are available at:  Our experience with the concordance studies has been the subject of several presentations and likely future publications.  Future studies will involve examining the ability of expert systems to decipher DNA mixtures similar to what would be encountered in forensic casework. In addition, NIST scientists also play an advisory role on the National Institute of Justice (NIJ) Expert System Testbed (NEST) project and have aided review of data obtained from forensic DNA laboratories with various expert system software programs.


Publications or Presentations Resulting From This Project:

Becky Hill presentation at the Expert Systems Workshop held in conjunction with the 2nd Annual Present and Future Technological Advances in Human Identification Conference (Roanoke, VA), March 27, 2006, "NIST Experience with FSS-i3 software" [.pdf]

Amy Decker presentations for Promega Technology Tour (Fairfax, VA), June 13, 2006, (Hartford, CT), June 15, 2006, and (Phoenix, AZ), June 22, 2006, "NIST Experience with FSS-i3 Software" [.pdf]

Becky Hill presentations for Promega Technology Tour (Austin, TX), June 13, 2006, (Chicago, IL), June 15, 2006, (Anaheim, CA), June 20, 2006, "NIST Experience with FSS-i3 Software" [.pdf]

Becky Hill presentation at 17th International Symposium on Human Identification (Nashville, TN), October 11, 2006, "NIST Experience Using v4.1.3 of FSS-i3 Software" [.pdf]

Becky Hill presentation at Fourth Annual Advanced DNA Technical Workshop (Bode) West (San Diego, CA), April 10, 2007, "NIST Experience with FSS-i3 v4.1.3 Software"

Amy Decker presentation at Fourth Annual Advanced DNA Technical Workshop (Bode) East (Captiva Island, FL), May 21, 2007, "NIST Experience with FSS-i3 v4.1.3 Software"

[Return to NIJ Projects page] [Return to STRBase]

Last updated: 06/27/2007

Disclaimer: This project was supported by National Institute of Justice Grant Number 2003-IJ-R-029, which is an interagency agreement between NIJ and the NIST Office of Law Enforcement Standards, awarded by the National Institute of Justice, Office of Justice Programs, US Department of Justice. Points of view in this document are those of the authors and do not necessarily represent the official position or policies of the US Department of Justice. Certain commercial equipment, instruments and materials are identified in order to specify experimental procedures as completely as possible.  In no case does such identification imply a recommendation or endorsement by the National Institute of Standards and Technology nor does it imply that any of the materials, instruments or equipment identified are necessarily the best available for the purpose.