News & Updates
New Y-STR interpretation guidelines for Germany
Two articles (in German) describe how the YHRD is used to interprete Y-STR results in forensic practice, see
Willuweit, S., Anslinger, K., Bäßler, G. et al. ‘Joint recommendations of the project group “Statistical analysis of DNA” and the German Stain Commission on the statistical analysis of Y‑chromosomal DNA typing results’ Rechtsmedizin (2018)
11th Haploid Markers Conference “Inferring Ancestry from DNA”
To attend the Haploid Markers 2018 workshop at 17-19th May 2018 in Bydgoszcz and to present a talk and/or poster, please register on-line at HM2018(posted almost 2 years ago)
Yet another huge update of the YHRD. More than 9000 Yfiler, 5000 PPY23 and almost 4000 YfilerPlus haplotypes have been added (see release history).
As a major change, we’ve updated the Discrete Laplace haplotype frequency estimation method as follows:
- DL is calculated for minimal and Yfiler haplotype resolution only.
- The static cut-off of 1.0e-08 was replaced by a metapopulation-specific cut-off described in a publication soon to be published. To give a quick glimpse: We apply a certain distribution to the histogram of DL estimates of all singleton haplotypes found in a particular metapopulation and take the 50% quantile as the estimation limit (cut-off).
- Please see the updated list of feasible metapopulations for an overview on the availability of DL.
Additionally, we implemented a neighbor search (+/- one allelic mutation step per locus) accessible for everybody as a new ancestry information feature.(posted almost 2 years ago)
Today we added more than 4,500 haplotypes from diverse destinations as Hiva Oa in French Polynesia, Zabol in Iran or Bodrogköz in Hungary. The database has now 188,209 haplotypes from 1,099 different populations in 132 countries. For details see the release history.(posted about 2 years ago)
Bug-Fix Kinship Tool
We’ve fixed an error in determination of the haplotype frequency in Kinship tool. Unter certain circumstances an actual observed haplotype was treated as not-observed and therefore estimated with a frequency of 1/N+1.(posted over 2 years ago)