probability of drop-in

58 views

Skip to first unread message

Aelbrecht Charlotte

unread,

Feb 4, 2019, 5:36:10 AM2/4/19

to lrmix-studi...@googlegroups.com

Hello,

We have a little discussion between experts on how to calculate the probability of drop-in that we use in LRmixStudio.

In the ISFG guidelines ‘DNA commission of the International Society of Forensic Genetics: Recommendations on the evaluation of STR typing results that may include drop-out and/or drop-in using probabilistic methods’ we can find the following:

Negative controls can be used: If n negative controls are analysed and there are x spurious alleles observed, then Pr(C)is estimated as x/n.

Because the pr(C) in LrmixStudio is used per locus and for me means Pr(C)=0.05 corresponds to 1 expected allele drop-in per 20 loci, don’t we have to dived by the numbers of loci? So for a 16 multiplex it would be Pr (C)= x/(n*16).

It depends on how it is programmed in LRmixStudio, so I hope you can help us in this debate.

With kind regards,

Charlotte

Charlotte Aelbrecht, ir. | Gerechtelijk Deskundige DNA

Tel +32 2 243 46 14 | GSM +32 477 99 58 34 | Fax +32 2 240 05 01

Nationaal Instituut voor Criminalistiek en Criminologie | Institut National de Criminalistique et de Criminologie

Vilvoordsesteenweg 100, 1120 Brussel | www.nicc.fgov.be

$Description: Description: Description: Description: Description: Description: C:\Users\ibuys\Desktop\new logo\Logo_NICC-kleur.gif$ Haalt veel uit weinig – Obtient beaucoup à partir de peu

hinda haned

unread,

Feb 10, 2019, 3:24:43 AM2/10/19

to Aelbrecht Charlotte, lrmix-studi...@googlegroups.com

Hello Charlotte,
To simplify things a bit, in LRmix Studio, each occurrence of drop-in per locus, is assigned Pr(C), and if there is no drop-in it assigns 1- Pr(C). Whether there is drop-in or not is determined by comparing the profiles under a given hypothesis.
Your question seems to be about how to determine the probability of drop-in in an experimental setting, you are correct in the way that you calculate Pr(C)=x/(n*L) where x= no of dropin events observed and n=number of samples and L=no of loci per sample. You can use this as the drop-in parameter in the software.
Hope this helps.
Hinda

--
You received this message because you are subscribed to the Google Groups "LRmix Studio user group" group.
To unsubscribe from this group and stop receiving emails from it, send an email to lrmix-studio-user...@googlegroups.com.
To post to this group, send email to lrmix-studi...@googlegroups.com.
To view this discussion on the web visit https://groups.google.com/d/msgid/lrmix-studio-user-group/bc1a20cafbc2456b9bbd0883123a4912%40just.fgov.be.
For more options, visit https://groups.google.com/d/optout.