Outgroup choosing for Outgroup F3 Statistics,about dreichlab/admixtools

bumblenick commented on July 20, 2024

Please send me your output (logfile). And if the ancient sample is a target you need diploid calls or more than one sample. Nick

…

On Sun, Mar 26, 2017 at 10:26 AM, hssKK ***@***.***> wrote: Hi, I have a ancient goat sample and want to know which modern goat group it most close to by using Outgroup F3 Statistics.But my outgroup had a long divergence time with them,so the F3_result is greater than 1.Does it may have some effect to my judge?Is my ancient goat most close to the modern goat group with the max F3_result? Thanks! — You are receiving this because you are subscribed to this thread. Reply to this email directly, view it on GitHub <#19>, or mute the thread <https://github.com/notifications/unsubscribe-auth/AQn_h6LNtlWOzSTgsW-DgQkn7ppXuNbsks5rpnWNgaJpZM4MpcW9> .

from admixtools.

biozzq commented on July 20, 2024

Dear @bumblenick

Outgroup-f3 test can be used to measure the shared genetic drift between two test populations. The f3 result is closely related to the selected outgroup. If the selected outgroup is far away from the test populations on genetic, I think the results maybe greater than 1. In human, if the outgroup is chimpanzee, the results maybe greater than 1. Is this right?

Best
Zhuqing

from admixtools.

hssKK commented on July 20, 2024

Dear @bumblenick
here is my result:
Source 1 Source 2 Target f_3 std. err Z SNPs
Ancient groupA Outgroup 1.196981 0.005023 238.302 3926773
Ancient groupB Outgroup 1.224042 0.005237 233.734 6541385
Ancient groupC Outgroup 1.247120 0.005399 230.995 5615777
Ancient groupD Outgroup 1.228669 0.005196 236.460 5276082
Ancient groupE Outgroup 1.200275 0.005038 238.225 6352217
Ancient groupF Outgroup 1.211333 0.005147 235.364 6104010
Ancient groupG Outgroup 1.217125 0.005233 232.604 5978688
Ancient groupH Outgroup 1.157842 0.004940 234.362 5602913
Is that mean my Ancient sample is most close to groupC ?
The results is greater than 1,does it may have some effect ?
What's more,because of the low coverage (1.1X ),heterozygosity is hard obtained from genotype calls ,does it may have some effect ?
I’m looking forward to your reply.
Thanks!

from admixtools.

bumblenick commented on July 20, 2024

You can't use f3 on a single sample unless you have good diploid calls which requires about 20X coverage. I wouldn't trust these results at all. N

…

On Mon, Mar 27, 2017 at 12:04 PM, hssKK ***@***.***> wrote: Dear @bumblenick <https://github.com/bumblenick> here is my result: Source 1 Source 2 Target f_3 std. err Z SNPs Ancient groupA Outgroup 1.196981 0.005023 238.302 3926773 <(302)%20392-6773> Ancient groupB Outgroup 1.224042 0.005237 233.734 6541385 <(734)%20654-1385> Ancient groupC Outgroup 1.247120 0.005399 230.995 5615777 Ancient groupD Outgroup 1.228669 0.005196 236.460 5276082 Ancient groupE Outgroup 1.200275 0.005038 238.225 6352217 <(225)%20635-2217> Ancient groupF Outgroup 1.211333 0.005147 235.364 6104010 <(364)%20610-4010> Ancient groupG Outgroup 1.217125 0.005233 232.604 5978688 <(604)%20597-8688> Ancient groupH Outgroup 1.157842 0.004940 234.362 5602913 Is that mean my Ancient sample is most close to groupC ? The results is greater than 1,does it may have some effect ? What's more,because of the low coverage (1.1X ),heterozygosity is hard obtained from genotype calls ,does it may have some effect ? I’m looking forward to your reply. Thanks! — You are receiving this because you were mentioned. Reply to this email directly, view it on GitHub <#19 (comment)>, or mute the thread <https://github.com/notifications/unsubscribe-auth/AQn_hzBt_p51d5uhW-Kg6tk0gmceKibbks5rp94LgaJpZM4MpcW9> .

from admixtools.

hssKK commented on July 20, 2024

Dear @bumblenick
It's hard for me to have an ancient sample which have coverage more than 20X . I also have two ancient sample ( mean coverage1.5X ) , if I take they as a population , is that enough ?
Or can I use D-statistic D(outgroup,ancient;A,B) to decide whilch group my ancient sample most close to ?
Or can you give my some suggestion on this problem?
Thanks a lot !

from admixtools.

bumblenick commented on July 20, 2024

With 2 samples you can use f3 (qp3Pop, inbreed: YES) N

…

On Tue, Mar 28, 2017 at 5:52 AM, hssKK ***@***.***> wrote: Dear @bumblenick <https://github.com/bumblenick> It's hard for me to have an ancient sample which have coverage more than 20X . I also have two ancient sample ( mean coverage1.5X ) , if I take they as a population , is that enough ? Or can I use D-statistic D(outgroup,ancient;A,B) to decide whilch group my ancient sample most close to ? Or can you give my some suggestion on this problem? Thanks a lot ! — You are receiving this because you were mentioned. Reply to this email directly, view it on GitHub <#19 (comment)>, or mute the thread <https://github.com/notifications/unsubscribe-auth/AQn_h_rjiQnTVBee0xTgPLiHkfbW8Iqyks5rqNhUgaJpZM4MpcW9> .

from admixtools.

hssKK commented on July 20, 2024

Thank you very much !

from admixtools.

Outgroup choosing for Outgroup F3 Statistics about admixtools HOT 7 CLOSED

Comments (7)

Related Issues (20)

Recommend Projects

React

Vue.js

Typescript

TensorFlow

Django

Laravel

D3

Recommend Topics

javascript

web

server

Machine learning

Visualization

Game

Recommend Org

Facebook

Microsoft

Google

Alibaba

D3

Tencent