Question: Batch effect problem using SVA showing problems
1
gravatar for amoltej
3.2 years ago by
amoltej10
Australia
amoltej10 wrote:

Hello everyone, 

I am working with 88 samples. To remove batch effect from the samples I am trying to use SVA package. But when I am trying to estimate number of latent factors. I can see in the tutorial and other questions on different forums,  the number of surrogate variables is really small. But I am estimating surrogate number - 83. When I am changing method from "leek" to "be" the number goes down to 1. When I try to estimate surrogate variables, it doesn't work.

I am not sure if these numbers are dependable or I am doing something wrong. Can somebody help me? 

Thanks in advance

sva batch effect combat sva • 566 views
ADD COMMENTlink modified 3.2 years ago by Jeff Leek560 • written 3.2 years ago by amoltej10
Answer: Batch effect problem using SVA showing problems
0
gravatar for Jeff Leek
3.2 years ago by
Jeff Leek560
United States
Jeff Leek560 wrote:

In this case, I would trust method="be". method="leek" tends to work better when the sample size is smaller (20-30 samples or less). 

Best

Jeff

ADD COMMENTlink written 3.2 years ago by Jeff Leek560

Hello, Thanx for the reply. May I know reason behind it? and how the big number for the latent factors is wrong?

 

Thanx

ADD REPLYlink written 3.2 years ago by amoltej10
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 16.09
Traffic: 400 users visited in the last hour