Values on Omega report

46 views
Skip to first unread message

Jaime Osorio

unread,
Aug 9, 2019, 9:16:22 AM8/9/19
to OmegaPlus
Hello, 

I'm trying to run Sweed and omega to discover the selective sweeps using a vcf file with 9,003 SNPs; however, the results of the Omega report are always zero, this is normal?, what indicates?. 

Thanks in advance

Jaime 

Nikolaos Alachiotis

unread,
Aug 12, 2019, 12:14:57 PM8/12/19
to OmegaPlus
Hello Jaime,
My guess is that the -maxwin parameter is not large enough to create regions with a sufficient number of SNPs.
Try to increase the -maxwin, and if the problem persists, please send me a sample of your vcf along with the command line to reproduce it.
Regards,
Nikos

--
You received this message because you are subscribed to the Google Groups "OmegaPlus" group.
To unsubscribe from this group and stop receiving emails from it, send an email to omegaplus+...@googlegroups.com.
To view this discussion on the web visit https://groups.google.com/d/msgid/omegaplus/ff344bc7-a667-4249-a0f8-86806c9267ec%40googlegroups.com.


--
Nikolaos Alachiotis

Jaime Osorio

unread,
Aug 13, 2019, 10:31:08 AM8/13/19
to omeg...@googlegroups.com
Nikos

Thanks for your response. I tried different -maxwin parameters and I obtained the same results (zero values). 
Command line: ./OmegaPlus -input file.vcf -grid 100 -name test -minwin 1000 -maxwin 10000 -seed  9382


I attached the vcf. 

Thanks 

Nikolaos Alachiotis

unread,
Aug 14, 2019, 12:14:10 PM8/14/19
to OmegaPlus
Hello Jaime,

The problem is the values you provide for -grid and -maxwin given the length in bp of your data and the number of SNPs in there.
Take for example chr1. The region size is about 37Mbp and there are 872 SNPs in there. The 100 grid points are distributed equidistantly over the 37Mbp, and up to 10000 bp on both directions of each point are searched. Which means that the number of SNPs found in the proximity of each grid point is very low, possibly 0 to a few. To start seeing scores and then decide how thoroughly you want to analyze your data simply increase -grid first, e.g., -grid 10000, and then give a larger to -maxwin.

Best regards,
Nikos






--
Nikolaos Alachiotis

Jaime Osorio

unread,
Aug 16, 2019, 8:30:37 AM8/16/19
to omeg...@googlegroups.com
Hi Nikos

Now works !!! thanks for your help

Jaime

Reply all
Reply to author
Forward
0 new messages