clustering samples

0

Entering edit mode

2.8 years ago

gs000095 ▴ 10

Hey all,

I have gene expression data for ~1000 samples. I am interested in a particular gene and I would like to see how these samples cluster based on this gene's expression. I was wondering :

If its correct to do it this way?
If yes, how should I perform the clustering? Should I leave the gene in the row and samples in the column and then perform hierarchical clustering ?

I look forward to your response and suggestions,

Best, SG

clustering • 681 views

ADD COMMENT • link updated 2.8 years ago by ATpoint 85k • written 2.8 years ago by gs000095 ▴ 10

2

Entering edit mode

A single gene? Wouldn't something simple like quantiles do here and then you simply check to which quantile the sample belongs? I mean, with only one gene you cannot really define clusters, do you?

ADD REPLY • link 2.8 years ago by ATpoint 85k

0

Entering edit mode

Mixture of distributions could do, if there is any underlying distribution

ADD REPLY • link 2.8 years ago by German.M.Demidov ★ 2.9k

Login before adding your answer.