Subject: RE: About propShared
From: Valeria Paccapelo
Date: Tue, 13 Oct 2009 15:22:11 +0100
To: "Jombart, Thibaut"

Dear Thibaut, 

Thanks for answering soon. Your example is really clear and it is in
concordance with what I thought the function does. 

I attach a file with an example. It consists on 7 individuals and 87
markers. In this example genotypes are homozygote only. 

For individuals 1 and 2,
- There are 4 markers which have NA for at least one of the individuals.
Then we have 87-4=83 markers to compare. 
- From these 83 markers, 56 are equal.
- The proportion of shared alleles is 56/83= 0.674698795180723
- but, propShared result is 0.6666667

Maybe, in this case the difference is no so important but in other cases the
difference is more important. Am I making any mistake when I calculate the
value manually?

Thanks, 

Yours, 

Valeria.

-----Mensaje original-----
De: Jombart, Thibaut [mailto:t.jombart@imperial.ac.uk] 
Enviado el: Martes, 13 de Octubre de 2009 10:59 a.m.
Para: Valeria Paccapelo
Asunto: RE: About propShared 

Dear Valeria, 

thanks for your interest in adegenet. Non-typed loci are discarded on a
pairwise basis during computations. For instance, if we have the following
genotypes:
#1: a/a c/d NA
#2: a/b d/d x/x
#3: NA NA x/x

propShared should give the following results:
1-2: 0.5
1-3: NA
2-3: 1

If you think your results are odd, can you please send me a reproducible
example?

Best,

Thibaut.


-- ###################################### Dr Thibaut JOMBART MRC Centre for Outbreak Analysis and Modelling Department of Infectious Disease Epidemiology Imperial College - Faculty of Medicine St Mary's Campus Norfolk Place London W2 1PG United Kingdom Tel. : 0044 (0)20 7594 3658 t.jombart@imperial.ac.uk http://www1.imperial.ac.uk/medicine/people/t.jombart/ http://adegenet.r-forge.r-project.org/ ________________________________________ From: Valeria Paccapelo [valeria.paccapelo@advantasemillas.com.ar] Sent: 13 October 2009 14:32 To: Jombart, Thibaut Subject: About propShared Dear Sir, I am using the function propShared of the R package adegenet which I found very useful. I am writing in order to ask you about how it works when there are non-typed loci for some individuals because I got some warnings and I do not find exactly the same values when I do it manually. Thanks in advance. Valeria Paccapelo. Advanta Semillas Balcarce Experimental Station CC 30 Ruta 226 Km. 60.5 (7620) Balcarce, Buenos Aires Argentina