Question: plink 1.9 Principal Component Analysis
0
gravatar for peris
6.0 years ago by
peris120
United States
peris120 wrote:

Hi all, 

I have done a plink pca test on my cas/control samples and generated *.eigenval and *.eigenvec files. Now how can I identify the outliers. I am not comfortable in using EIGENSOFT. Is there any other way to do the same.

 

Thanks and regards. 

plink pca • 5.8k views
ADD COMMENTlink modified 4.5 years ago by Biostar ♦♦ 20 • written 6.0 years ago by peris120

Hi, did you find out how to plot the eigenvec? I know you can use plot() but how can one add the sample labels?

ADD REPLYlink written 3.1 years ago by DataFanatic230
3
gravatar for chrchang523
6.0 years ago by
chrchang5237.1k
United States
chrchang5237.1k wrote:

You should be able to load the .eigenvec file in R, perform clustering on it, and identify outliers in the process.

(The old PLINK way to do this was the --neighbour command.  This was based on a basic identity-by-state matrix instead of the variance-normalized GRM, but results should be similar.)

ADD COMMENTlink modified 6.0 years ago • written 6.0 years ago by chrchang5237.1k

Hi chrchang523, thanks for your reply. I need some tip from you again. I know very little R and very very new in GWAS related studies. I can plot grap from the PCA using R; but how can I perform clustering for identifying the outliers.

ADD REPLYlink written 6.0 years ago by peris120
2

A large number of clustering routines have been written for R.  http://cran.r-project.org/web/views/Cluster.html is a reasonable starting point.

ADD REPLYlink written 6.0 years ago by chrchang5237.1k
Please log in to add an answer.

Help
Access

Use of this site constitutes acceptance of our User Agreement and Privacy Policy.
Powered by Biostar version 2.3.0
Traffic: 1206 users visited in the last hour