search

alachins / raisd

RAiSD: software to detect positive selection based on multiple signatures of a selective sweep and SNP vectors
33 stars 13 forks source link

window size, proxy for strength/timing of selection #31

Open cistarsa opened 2 years ago

cistarsa commented 2 years ago

Hello and thank you developing this nifty (and easy to implement) program! I've had success identifying swept regions in two divergent populations (n=25 & 28) and I'm wondering if I can infer the relative timing of the sweep using the window size, or muLD alone?

alachins commented 2 years ago

Hello and thank you for using RAISD. I do not know if and how this can be done using the window size or muLD. If you know the recombination rate of a region, perhaps you can use the corresponding alpha value of that region that SweeD reports (not RAiSD in this case). There is a relationship that relates alpha, the recombination rate, the selection coefficient, and the population size, from which you could calculate the selection coefficient.

On Wed, Dec 8, 2021 at 12:07 AM cistarsa @.***> wrote:

Hello and thank you developing this nifty (and easy to implement) program! I've had success identifying swept regions in two divergent populations (n=25 & 28) and I'm wondering if I can infer the relative timing of the sweep using the window size, or muLD alone?

— You are receiving this because you are subscribed to this thread. Reply to this email directly, view it on GitHub https://github.com/alachins/raisd/issues/31, or unsubscribe https://github.com/notifications/unsubscribe-auth/AALKWCTOEC2ZO5SX4K5ZJ6LUP2HRZANCNFSM5JSMVTWQ . Triage notifications on the go with GitHub Mobile for iOS https://apps.apple.com/app/apple-store/id1477376905?ct=notification-email&mt=8&pt=524675 or Android https://play.google.com/store/apps/details?id=com.github.android&referrer=utm_campaign%3Dnotification-email%26utm_medium%3Demail%26utm_source%3Dgithub.

-- Nikolaos Alachiotis

cistarsa commented 2 years ago

Great, thank you for the quick reply! Hm. Interesting, I do have N estimates from dadi and staiwayplot, but these are not window based or even chromosmal....would a population level estimates suffice for alpha=4Ns. I do have "scaffold" level recombination rates from pyrho. Yet it seems sweed estimates these parameters from input (vcf)? If that's enough than I can run sweed and compare with raisd, then estimate s...? (sorry for the train of thought response) And thanks again for your insight!

cistarsa commented 2 years ago

Hello again, I've been trying to determine common outliers with Sweed results of various grid sizes, but I'm receiving this error:

gsl: interp.c:37: ERROR: insufficient number of points for interpolation type

Here, I'm using a grid size of 10,000 on 153 scaffolds with an N50 of 15MB; is it possible to run Raisd without the grid option and compare windows with sweed using something like bedtools intersect to find shared CLR?

alachins commented 2 years ago

Hi. Sorry, I missed your previous post. The grid option is needed if you want to use the built-in option of RAiSD for the common-outlier analysis. You can try a small grid size first, e.g., 10 or 100, instead of 10,000. You can gradually increase it until RAiSD starts giving this error. Then, use this grid size (the max grid size for which RAiSD does not give the error) for SweeD. I hope this helps.

On Mon, Dec 13, 2021 at 6:33 PM cistarsa @.***> wrote:

Hello again, I've been trying to determine common outliers with Sweed results of various grid sizes, but I'm receiving this error:

gsl: interp.c:37: ERROR: insufficient number of points for interpolation type

Here, I'm using a grid size of 10,000 on 153 scaffolds with an N50 of 15MB; is it possible to run Raisd without the grid option and compare windows with sweed using something like bedtools intersect to find shared CLR?

— You are receiving this because you commented. Reply to this email directly, view it on GitHub https://github.com/alachins/raisd/issues/31#issuecomment-992709497, or unsubscribe https://github.com/notifications/unsubscribe-auth/AALKWCUHOCQ6YHZOJ3NTHA3UQYU7XANCNFSM5JSMVTWQ . Triage notifications on the go with GitHub Mobile for iOS https://apps.apple.com/app/apple-store/id1477376905?ct=notification-email&mt=8&pt=524675 or Android https://play.google.com/store/apps/details?id=com.github.android&referrer=utm_campaign%3Dnotification-email%26utm_medium%3Demail%26utm_source%3Dgithub.

-- Nikolaos Alachiotis

cistarsa commented 2 years ago

no worries at all! Thanks that's helpful and is there a way to iteratively test the grid size or a lower limit? Still getting this error at -G 5.