Lutz Roewer, Sascha Willuweit Dept. Forensic Genetics, Institute of Legal Medicine and Forensic Sciences Charité Universitätsmedizin Berlin, Germany

Transcription

1 The new YHRD Lutz Roewer, Sascha Willuweit Dept. Forensic Genetics, Institute of Legal Medicine and Forensic Sciences Charité Universitätsmedizin Berlin, Germany

2

3 Aug 99 Jun 00 Jan 03 Jul 03 Aug 03 Oct 03 Oct 03 Dec 03 Feb 04 Feb 04 Apr 04 May 04 Jul 04 Sep 04 Dec 04 May 05 Oct 05 Jan 06 Aug 06 Dec 06 Apr 07 Aug 07 Jul 08 Oct 08 Jan 09 Feb 09 Mar 09 Jun 09 Aug 09 Nov 09 Dec 09 Mar 10 Jul 10 Dec 10 May 11 Jun 11 Dec 11 Feb 12 Aug 12 Oct 12 Jan 13 Jan 13 Jul 13 Oct 13 Dec 13 Aug 14 Nov 14 Feb Growth rates Minimal Haplotype Yfiler Powerplex Y

4 Release 49 (February 2015): 143,044 haplotypes

5

6 128 countries

7 China: 19,589 USA: 7,793 Spain: 7,775 Germany: 7,675 Brazil Poland: Average: 152 Micronesia: 19 DR Kongo: 13 Solomon Islands: 8

8 National Databases

9 Representation of continents

10 Metapopulation assignment three MP levels

11 Details on submissions

12 Datasets

13 !!!

14 Y-SNP and haplogroup analysis

15 What s new? Free admission, no registration, no tracking Search facilitated using input files (Excel, XML, CSV, Genemapper) allowing bulk searches Reduction of the output files: frequency estimation in 5 differently resolving datasets Add-on features can be selected (frequncy estimates in selected national or metapopulation databases, ancestry information (AI) on the searched haplotype) No assignment of high-resolution haplotypes to geopositions (misleading information!) Different frequency estimation methods (constant and variable) Tools (Mixture, Kinship, AMOVA) SNP/Haplogroup designation based on the updated Phylotree Y (van Oven et al.) Fast, principle of fair use YHRD has full version history

16 Search options

17 Import large Excel files (>100, <200)

18 Frequency values for two selected profiles

19 Jump between datasets

20 Jump between datasets

21 Intelligent counting: use the most informative haplotype format in the largest available database 27 loci Count 0/19,593 Frequency 5.1 x loci 0/71, x loci 260/125, x 10-3

22 Frequency calculation (beyond the limits of the database size)

23 What is the evidentiary value of a haplotype match Match probability is the probability, that an individual, which is randomly selected from the population, has the same DNA profile as the DNA profile which was found at the crime scene If we would know all haplotypes in a population, than the frequency of the haplotype in question would be the match probability But because we don t know the true frequencies, we need to estimate the frequencies to calculate the match probability

24 Frequency estimation methods Constant estimators Variable estimators Augmented counting (1/n+1) with CI Counting with database inflation (Brenners κ) Surveying method (Krawczak) Coalescence based estimation (Caliebe) Laplace method (Andersen)

25 Example for rare haplotype calculation using the kappa inflation (in case of 0 observations) Count singletons using kappa 23 loci impossible 17 loci x 10-6 K=0.78 (1.4 x 10-5 )* 9 loci x 10-6 K=0.24 (7.9 x 10-6 )* * Counting

26

27 Fig. 1 Comparison of (1) the relative frequency of a haplotype (number of times it has been observed divided by the database size) and (2) the estimated haplotype frequency using the discrete Laplace method. Note, that for frequently observed haplotypes, t... Mikkel Meyer Andersen, Poul Svante Eriksen, Niels Morling Cluster analysis of European Y-chromosomal STR haplotypes using the discrete Laplace method Forensic Science International: Genetics, Volume 11, 2014,

28 Variable estimators Coalescence vs. all Discrete Laplace vs. all Courtesy of M.M. Andersen (Copenhagen)

29 Tools of the YHRD Mixture calculation (frequency and LR based) Population differentiation (AMOVA, Fst/R st, MDS) Kinship calculation (frequency and LR based)

30

31 Mixture (Import or type)

32 Wolf A., Caliebe A., Junge O., Krawczak M. (2005), Forensic interpretation of Y-chromosomal DNA mixtures., Forensic Sci Int 152(2-3),

33 Likelihood calculation for patrilinear relatives (Example) Brothers? Mutation: µ (DYS389II) = 3.6 x , 13, 30, 24, 11, 13, 13, f obs = 1.4 x 10-4* 14, 13, 31, 24, 11, 13, 13, f obs = 2.3x10-5* Meioses * YHRD L(X) = 1.4 x 10-4 x 1 x µ/ x 10-5 x 1 x µ/2 (related) = 2.9 x 10-7 L(Y) = 1.4 x 10-4 x 2.3 x 10-5 (non-related) = 3.2 x 10-9 LR (X/Y) = 91 for related Charité Universitätsmedizin Berlin 2015

34 Kinship (Import or type) Buckleton JS., Triggs CM., Walsh SJ. (2005), Forensic DNA evidence interpretation, 1st ed., CRC press Rolf B, Keil W., Brinkmann B., Roewer L., Fimmers R. (2001), Paternity testing using Y-STR haplotypes: assigning a probability for paternity in cases of mutations., Int J Legal Med. 115, 12-5

35

36

37 Test on population substructure (Fst, Rst) (Example: 17,278 Chinese individuals in 52 populations)

38 Ancestry information for a searched haplotype

39 For ancestry information each haplotype will be reduced to the core (9 loci)

40 Three features can be checked-out 1 2 3

41 Heat map (based on the minimal haplotype core set, which is part of each submitted haplotype, best possible representation) High resolution haplotypes are not longer locatable!

42 FAQ

43 Training

44

45

46 Summary YHRD is growing every month, making the Y chromosome (besides mtdna) the best described human polymorphism used in forensic genetic analysis YHRD accomodates the profiles of the new generation Y-STR kits YHRD includes constant and variable estimators to evaluate the match probability of frequent and rare haplotypes YHRD provides interpretative tools (AMOVA&MDS, Mixture, Kinship)

47 Let s continue to discuss YHRD at the 10th Y Chromosome User Workshop in Berlin, Germany Spring 2016