Question: Tac and Bioconductor giving different results
gravatar for kritikamish99
13 months ago by
kritikamish9910 wrote:


The questions may seem very general please bear!

I have done analysis using TAC affymetrix software and bioconductor affy package

the result are not showing similarity between both

Which method i should go for TAC or Bioconductor

ADD COMMENTlink modified 13 months ago by Steve Lianoglou12k • written 13 months ago by kritikamish9910
gravatar for Steve Lianoglou
13 months ago by
Steve Lianoglou12k wrote:

This is an impossible question to answer.

Although I (and likely most people here) have never used "TAC affymetrix software" (I guess you mean this?), I imagine there are a multitude of ways to perform an analysis there just as there are many ways to perform an analysis using various bioconductor tools.

You haven't specified what you've done exactly, so it's impossible to say which you should go with (for all we know, both analyses could have been done incorrectly).


ADD COMMENTlink written 13 months ago by Steve Lianoglou12k

Hi Steve

I was performing u133a2 affymetrix gene expression study

The normalisation method which i used was MAS5

After that i did FC calculation manually and then logFC manually too over exprs data that is eset i am getting different results compare to TACs .

For example in column i am getting FC value as 2 but for same probe in TAC it is showing -8. This Should not be so different


ADD REPLYlink written 13 months ago by kritikamish9910

You are likely right, you shouldn't be seeing such large differences. Again, no one can help you because we can't debug stories -- you need to provide code.

You probably want to RMA normalize your data and use limma for analysis. You hopefully have some intuition of the experiment you are analyzing, so look at some positive control genes you expect to go up and down and see if you get the expected logFC's.

If you are still having problems, you need to update your original question by providing a little more information about your experimental design and copy and pasting the code you used to conduct your analysis. Including a portion of your "targets" file that shows the group mapping to each sample would be helpful as well.

ADD REPLYlink written 13 months ago by Steve Lianoglou12k

Hi Steve

I was able to reproduce same results from both the methods

Actually the out put file which i got from bioconductor was logged and TAC gives FC as unlogged after converting to antilog to my data i was able to get same results


ADD REPLYlink written 13 months ago by kritikamish9910
Please log in to add an answer.


Use of this site constitutes acceptance of our User Agreement and Privacy Policy.
Powered by Biostar version 2.2.0
Traffic: 474 users visited in the last hour