cuffdiff for groups

Question

[Vasu]

I would like to know whether the following command is right or not?

cuffdiff -o diff_out4 -b ../genome/ce10.fa -p 2 -L em,la -u merged_asm/merged.gtf cond1.bam1,cond1.bam2,cond1.bam3(space)cond2.bam1,cond2.bam2,cond2.bam3

cond1.bam1,cond1.bam2,cond1.bam3 is group1

cond2.bam1,cond2.bam2,cond2.bam3 is group2

so basically I want group1 vs group2

Some told me that there should be only two conditions like the following:

cuffdiff -o diff_out4 -b ../genome/ce10.fa -p 2 -L larval,early -u merged_asm/merged.gtf cond1.bam cond2.bam

here cond1.bam is the merged bam file of the first three

cond2.bam is the merged bam file of the other three

But If I merge the bam files of samples in the group and from the output If i need to know the expression levels of the individual samples, I guess its not possible.

Answer 1

[Devon Ryan]

As I said on SEQanswers, cond1.bam1, cond1.bam2, and cond1.bam3 ARE NOT BIOLOGICAL REPLICATES! They're technical replicates. By lying to cuffdiff and telling it that you have biological replicates when, in fact, you don't you're going to get incorrect results.

[Vasu]

Yes @Devon. you are right. But If I do that how can I get the expression of individual replicate?

[Devon Ryan]

You can't, but why would you want to?

[Vasu]

let's say I'm not sure whether we collected the sample from the correct tissue. so we want to know if the expression is similar to the other ones in the group. so we need the individual expression profiles

[Devon Ryan]

I take it we're not talking about the dataset that you referenced on SEQanswers anymore. In any case it would make more sense to just use cuffnorm for that.

[Vasu]

ok. I would like to know d differnce between cuffnorm and cuffdiff?

[Devon Ryan]

From the manual:

Cufflinks includes a program, “Cuffnorm”, that you can use to generate tables of expression values that are properly normalized for library size.

Cufflinks includes a program, “Cuffdiff”, that you can use to find significant changes in transcript expression, splicing, and promoter use.

[Vasu]

Thank you !!