search

DReichLab / AdmixTools

Tools test whether admixture occurred and more
182 stars 64 forks source link

The Outgroup f3 statistic result is inconsistent with the NJ tree #21

Open ghost opened 7 years ago

ghost commented 7 years ago

Hi,

I have two ancient samples (7X and 13X) . In NJ tree , they get together with groupA . But in Outgroup f3 statistic ,the groupB's f_3 value is higher than gourpA ,How to explain the result ?

The NJtree may like this: http://i2.muimg.com/4851/d715cb8fb65cd020.png

The output : result: Ancient groupB outgroup 0.235691 0.001944 121.244 12220407 result: Ancient groupA outgroup 0.231413 0.001914 120.894 11370158

Thanks!

bumblenick commented 7 years ago

Can't help. The 2 calculations are (of course) quite different.

Nick

On Wed, Apr 5, 2017 at 8:00 AM, hisakawakuro notifications@github.com wrote:

Hi,

I have two ancient samples (7X and 13X) . In NJ tree , they get together with groupA . But in Outgroup f3 statistic ,the groupB's f_3 value is higher than gourpA ,How to explain the result ?

The NJtree may like this: http://i2.muimg.com/4851/d715cb8fb65cd020.png

The output : result: Ancient groupB outgroup 0.235691 0.001944 121.244 12220407 result: Ancient groupA outgroup 0.231413 0.001914 120.894 11370158

Thanks!

— You are receiving this because you are subscribed to this thread. Reply to this email directly, view it on GitHub https://github.com/DReichLab/AdmixTools/issues/21, or mute the thread https://github.com/notifications/unsubscribe-auth/AQn_hzLZVk8kBBBVTx9xR_O7vLtUvSiJks5rs4JNgaJpZM4M0I5j .

ghost commented 7 years ago

What seems more plausible ? if I want to see which group my ancient samples more like.

wlz0726 commented 7 years ago

this may help you understand the outgroup-f3 http://gaworkshop.readthedocs.io/en/latest/contents/06_f3/f3.html @hisakawakuro