Validation algorithm.

[tambonis]

Hi.

I work in the development of algorithm for work with differential expression with data from mRNA-Seq.
Now I need validate this algorithm. What should I do?


Thanks.

[danwiththeplan]

You might need to be a bit more specific about what programs you are using and what your algorithm does. If you need some test datasets of RNA sequencing, there are plenty available at the sequence read archive at NCBI.

[tambonis]

Sorry for the long delay.

I am using the edgeR/DESeq/baySeq tools of the R platform to rank genes involved in differential expression. I am also developing a method to rank genes involved in differential expression.

So, I need a count table of short reads and a list of true differentially expressed genes coming from RNA-Seq that relates two experimental different conditions.

[dpryan]

It might make the most sense to use the RNAseq samples from MAQC. That's the approach taken in many papers (I think they did that in the cufflinks/cuffdiff paper).