From hmviitutu.fi Sat Mar 24 11:33:23 2012
From: Heidi Viitaniemi <hmviitutu.fi>
To: Multiple Recipients of <crimap-usersanimalgenome.org>
Subject: Memory and hap_sys problems in crimap
Date: Sat, 24 Mar 2012 11:33:23 -0500
HI,
I'm working with SNP data on a F1 half-sib family with 20 offspring from
each sire (total 30 sires). Parents and offspring are genotyped for 10 000
SNPs. I would like to build a map for these, but so far I'm stuck with
memory problems and hap_sys in v2.503.
For the memory part I've been trying to adjust the settings on defh.s up
and down from what they are there with varying degree of success: I get a
warning from glibc with something corrupted and to me this looks more like
a problem of the linux than crimap itself. I tried to solve this by
removing and compiling crimap again, but I get the same problem still. I
get over this problem whendoing twopoint only for 1 chromosome at a time,
but I would still like to able to do twopoint for the whole data.
When dealing with only 1 chromosome worth of SNPs the twopoint goes fine,
but when I start building based on the twopoint I get an error for "Check
hap_sys0 systems for intralocus recombinants". Reading the crimap tutorial
and documentation hap_sys kind of makes sense to me, but I'm still a bit
unsere how the get a around it or where the information of the primary and
secondary haplotype systems can be found. I havent specified any hap_sys
systems in the .par file, but the hap_sys in the
.par file is 1. Any good suggestions on how to get over this problem?
All help is wellcome at this point =)
Thanks, Heidi Viitaniemi
--
______________________________________________
Heidi Viitaniemi PhD student Division of Genetics and Physiology Department
of Biology (Vesilinnantie 5) University of Turku 20014 Turku
tel: +358-2-3335808
|