search

Shuhua-Group / AdmixSim2

forward-time simulator of population genetic data
GNU General Public License v3.0
6 stars 3 forks source link

Unexpected behavior! #2

Open drdna opened 2 weeks ago

drdna commented 2 weeks ago

Running the python version of this script using a .hap file with 42621 sites (snv file has 42621 lines), produces output haplotypes with 49028 sites! It's creating variant positions out of thin air!

Model looks like this:

*1-10 Anc1,Anc2,Admix1 100 0.5,0.5,0 10000 0,0,1 10000 0,0,1 10000 0,0,1 10000 0,0,1 10000 0,0,1 10000 0,0,1 10000 0,0,1 10000 0,0,1 10000 0,0,1

Command looks like this:

python AdmixSim2.py -i AdmixSim2 -p Admix1 -g 10 -n 100 --mut-rate 0.000001 --rec-rate 0.0000006 -o AdmixSim2.out

Zhang-Rui-2018 commented 2 weeks ago

Hi! Thank you for using our software! Can I ask how long is the chromosome you are simulating?

drdna commented 2 weeks ago

Thanks for your speedy response. The chromosome is 3.66 megabases long but the number of polymorphic sites in the hap file is 42,621.

Zhang-Rui-2018 commented 2 weeks ago

Hi! It makes sense since you have set the mutation rate as 0.000001, the chromosome length as 3.66Mb, so the expectation new mutations should be chromosome length (in bp) * mutation rate * generation * #output haplotype, which is about 7k in your setting.

If you don't want new mutations, you could set the mutation rate as 0 or --no-mut in cpp version.

drdna commented 2 weeks ago

Now I see what’s happening. The program, has created a new .snv file too. I was expecting it to simply flip ones and zeroes but the actual behavior makes total sense.

drdna commented 2 weeks ago

Oh, I see what’s happening. The program, has created a new .snv file too. I was expecting it to simply flip ones and zeroes but the actual behavior makes total sense. I have changed the title of the issue to reflect my misunderstanding. I have also added some suggestions for added functionality.

Mark L. Farman Professor, Department of Plant Pathology 225 Plant Science Building 1405 Veteran's Dr. University of Kentucky Lexington, KY 40546 USA tel: (859) 218-0728 fax: (859) 323-1961

On Nov 8, 2024, at 12:28 AM, Rui Zhang @.***> wrote:

CAUTION: External Sender

Hi! It makes sense since you have set the mutation rate as 0.000001, the chromosome length as 3.66Mb, so the expectation new mutations should be chromosome length (in bp) mutation rate generation * #output haplotype, which is about 7k in your setting.

If you don't want new mutations, you could set the mutation rate as 0 or --no-mut in cpp version.

— Reply to this email directly, view it on GitHubhttps://github.com/Shuhua-Group/AdmixSim2/issues/2#issuecomment-2463795408, or unsubscribehttps://github.com/notifications/unsubscribe-auth/AAYUUDJKK4NQNVTYVFXRMLLZ7RDWHAVCNFSM6AAAAABRLXY3POVHI2DSMVQWIX3LMV43OSLTON2WKQ3PNVWWK3TUHMZDINRTG44TKNBQHA. You are receiving this because you authored the thread.Message ID: @.***>