1. UNCLASSIFIED Defense Forensic Science Center DNA Mixture Interpretation Study: Inter- and Intra-laboratory Variation Roman Aranda IV 1, Emily Rogers 1, Karl Mereus 1, Phillippa Spencer 2, and Rick Tontarski 1 Office of the Chief Scientist ASCLD, 28 April 2015 1 2

2. UNCLASSIFIED Disclaimer The opinions or assertions contained herein are the private views of the author and are not to be construed as official or as reflecting the views of the Department of the Army or the Department of Defense. Names of commercial manufacturers or products included are incidental only, and inclusion does not imply endorsement by the authors, DFSC, OPMG, DA or DoD. Unless otherwise noted, all figures, diagrams, media, and other materials used in this presentation are created by the respective author(s) and contributor(s) of the presentation and research. 2

3. UNCLASSIFIED Briefing Overview 3 Mixture Study Structure Preliminary Results Variation Illustrated via GIM and AM Path Forward: DNA Examiner Assessment Tool (DEAT)

4. UNCLASSIFIED Purpose: To assess the inter- and intra-laboratory variation in DNA examiners’ generated genotype interpretations To better understand the current state and potential limitations of mixture interpretation in the forensic community 4 DFSC Mixture Study

5. UNCLASSIFIED Initiated Summer 2014 183 examiners 55 forensic laboratories 5 Participation:

6. UNCLASSIFIED Examiners asked to interpret 6 mixtures: All examiners received identical 6 mixtures Used their laboratory’s SOP Stochastic and analytical thresholds set by DFSC Genotypes recorded on Excel-based worksheet provided User assessment form 6 Study Datasets:

7. UNCLASSIFIED 7 Metrics: GIM + AM How many answers did I provide at each locus? Did my genotypes include the “correct answer”? KnownGeneratedATMAFMInc 11, 12 200 11, Any100 11, 13110 10, 13020 Inc.002 Genotype Interpretation Metric: GIM Allelic Match: 00.5 1 InconclusiveExact Match Combinations + Anys

8. UNCLASSIFIED Automated program analyzes returned genotypes Uses worksheet template Calculates GIM/AM metrics 8 DEAT: DNA Examiner Assessment Tool

9. UNCLASSIFIED 9 GIM Box Plots = Median GIM Locus, Mixture, etc. = 25–75 Percentile = Statistical Max/Min Distribution = Outliers

10. UNCLASSIFIED 10 Mixture 1: 3.5:1, 2-person GIM Major and Minor Combined Region Local State Federal Int’l/Other

11. UNCLASSIFIED 11 Mixture 1: 2-person, 3.5:1 AF % AT % Median Method 1 Method 2 All Examiners Plotted Individually Major and Minor Combined

12. UNCLASSIFIED 12 Mixture 2 & 3: 2:1, 2-person GIM 2:1, w/ Ref2:1, w/o Ref B C D E F Method 1 2 B C D E F Method 1 2 Lab A Major and Minor Combined

13. UNCLASSIFIED 13 Inc % AT % Mixture 2 & 3: 2:1, 2-person 2:1, w/ Ref2:1, w/o Ref Major and Minor Combined Median Method 1 Method 2 Median Method 1 Method 2

14. UNCLASSIFIED 14 Mixture 5: 3-person, 4:1:1 w/ Ref GIM Method 1 Method 2 B C D E F Lab A

15. UNCLASSIFIED 15 Mixture 5: 4:1:1 w/ ref (Major) AF % AT % Method 2 Method 1 Major and Minors Combined

16. UNCLASSIFIED 16 Mixture 6: Box Plot GIM Method 1 Method 2 B C D E F Lab A

17. UNCLASSIFIED 17 Mixture 6: 3-person, 1:1:1 AF % AT % Method 2 Method 1 Major and Minors Combined

18. UNCLASSIFIED 18 Final Stat: 3.5:1, 2-person, Major Lab A Lab B 5 25 Examiner ~75% of examiners reported log (fs)

19. UNCLASSIFIED log (fs) 19 Final Stat: 2:1, 2-person, Major Contributor Lab A Lab B Lab A Lab B 55 20 w/ Ref w/o Ref 20 Examiner ~75% of examiners reported ~50% of examiners reported

20. UNCLASSIFIED 20 Final Stat: 4:1:1, 3-person w/ Ref, Major Lab A Lab B 0 25 ~55% of examiners reported Examiner log (fs)

21. UNCLASSIFIED Laboratory Use of DEAT 21

22. UNCLASSIFIED 22 Benchmarking GIM Mixture 1 2 3 1.Interpretation rate 2.Variation 3.Deconvolution

23. UNCLASSIFIED 1.Within Δ range 2.Minor Δ: review/retest? 3.Major Δ: retrain? 23 Examiner Assessment GIM Training/Proficiency Tests 2 3 1 Training of new and existing DNA examiners:

24. UNCLASSIFIED 24 Examiner Assessment AF% AT% or GIM 1.Within Δ range 2.Minor Δ: review/retest? 3.Major Δ: retrain? 2 3 1 Simulate whether those errors would still produce a match/hit Training of new and existing DNA examiners:

25. UNCLASSIFIED Summary Large variation in examiner generated genotypes: Within and between laboratories Variation in interpretation methods Creation of DNA Examiner Assessment Tool: quantify and assess changes via GIM and AM Potential use of metrics in DNA laboratories Benchmarking, training, and proficiency tests 25

26. UNCLASSIFIED Roman Aranda IV Research Scientist Office of the Chief Scientist CTR, A-T Solutions roman.aranda3.ctr@mail.mil 26