Our results are too noisy

[aguilar-gomez]

We would like to make our results less jumpy. We see there is an optional tolerance parameter for the convergence. We do not understand very well how to modify it. Can you please explain a little? What is the default parameter for this?

Thank you

[russcd]

I would be happy to help. Could you provide a bit more information?

Are you finding unstable estimates of admixture time or is the local ancestry across samples "jumpy"? Or is it something else?

[aguilar-gomez]

Hi,It is the local ancestry across samples that is jumpy. We are looking at very recently admixed samples and some of them we know they are F1 from the two parental populations and we still get a very noisy estimate accesos the genome. For the admixture time we actually also find a much older estimate than what we know is the admixture time. Thank you,Diana On Sep 17, 2023, at 8:07 AM, Russell Corbett-Detig ***@***.***> wrote: I would be happy to help. Could you provide a bit more information? Are you finding unstable estimates of admixture time or is the local ancestry across samples "jumpy"? Or is it something else? —Reply to this email directly, view it on GitHub, or unsubscribe.You are receiving this because you authored the thread.Message ID: ***@***.***>

[russcd]

Could you provide me with a sample file (e.g., one chromosome from the analysis you are trying to do) and the command line you are using for this?

[lin0yuan]

Hi Russell, I am the one who does the run for Diana. Here are the files and codes:
ancestry_hmm -i subset_ahmm_input_onechr -s sample_ploidy4ahmm -a 2 0.80 0.20 -p 0 100000 0.8 -p 1 -10 0.2 -e 0.01 --tmax 100 --pmax 1
The attached ahmm input file is the largest chromosome from our original ahmm input. To prepare the ahmm input, we used your provided vcf2ahmm.py to process our vcf with the parameters "--min_total 2 -m 0 --min_diff 0.1".
Thanks a lot for your help!
subset_ahmm_input.zip