Ignoring individuals in EIGENSOFT in IND file

[muhammadsohailraza]

Dear All,
I am trying to run Eigensoft's SMARTPCA module for my PCA analysis on world-wide populations of 1000 G together with my own samples. However, when i run SMARTPCA with remove outlier options, it removes some samples from 1000G populations and from my data set.

So, I run the SMARTPCA again (without outlier removal option) while ignoring the some of the outlier individuals in the "*.ind" file. as:
HG04239 M SAS
NA06984 M ???
NA06985 F ???
NA06986 M ???
NA10847 F CEU
NA10851 M CEU
NA11829 M CEU
NA11830 F CEU

But the E-values still calculated in the resultant output files for the individuals marked. Can anyone kindly guide me how to format the "ind" file to ignore some individuals from subset of populations?

Thank you very much!

--sohail

[bumblenick]

The population label should be "Ignore" (no quotes)
Also convertf with such a label will strip out such individuals.
In addition, if you specify the populations that you want to use
to make axes, (poplistname:) the ??? individuals will appear in the
evec file but this seems likely to be completely harmless.

Nick

On Tue, Nov 22, 2016 at 2:42 AM, SOHAIL [email protected] wrote:

Dear All,
I am trying to run Eigensoft's SMARTPCA module for my PCA analysis on
world-wide populations of 1000 G together with my own samples. However,
when i run SMARTPCA with remove outlier options, it removes some samples
from 1000G populations and from my data set.

So, I run the SMARTPCA again (without outlier removal option) while
ignoring the some of the outlier individuals in the "*.ind" file. as:
HG04239 M SAS
NA06984 M ???
NA06985 F ???
NA06986 M ???
NA10847 F CEU
NA10851 M CEU
NA11829 M CEU
NA11830 F CEU

But the E-values still calculated in the resultant output files for the
individuals marked. Can anyone kindly guide me how to format the "ind" file
to ignore some individuals from subset of populations?

Thank you very much!

--sohail

—
You are receiving this because you are subscribed to this thread.
Reply to this email directly, view it on GitHub
#14, or mute the thread
https://github.com/notifications/unsubscribe-auth/AQn_h0gRI_oJv4NFQoMJOVhPHuImQh_Jks5rApzQgaJpZM4K5GIr
.

[muhammadsohailraza]

issue resolved.. thank you @nick