Cuffdiff timecourse replicates

[Chinney14]

I am trying to figure out if CuffDiff handles replicates in its timecourse option.

so it would be something like this...

cuffdiff -T sampleA_20min_a.bam,sampleA_20min_b.bam,sampleA_20min_c.bam sampleA_40min_a.bam,sampleA_40min_b.bam,sampleA_40min_c.bam sampleA_1hr_a.bam,sampleA_1hr_b.bam,sampleA_1hr_c.bam

and, of course, I would include the other options. But basically, can I use a space to separate time points and commas to separate replicates?

Thanks in advance.

[dpryan]

Sort of, but that's not really a time-series analysis, cuffdiff will simply treat those as different groups. Cuffdiff isn't the most flexible package, you'd be better served with limma, DESeq2, or edgeR.

[Chinney14]

dpryan - I truncated the input for simplicity sake... I have the same treated/untreated cell line harvested in triplicate at 5 different time points. Do you have a preference for one of those for something like this?

you'd be better served with limma, DESeq2, or edgeR.

I appreciate the response.

[dpryan]

Since these aren't the exact same sample extracted at multiple time-points, I would normally use DESeq2. Others will have their own preferred tool.

[danwiththeplan]

Sort of, but that's not really a time-series analysis, cuffdiff will simply treat those as different groups. Cuffdiff isn't the most flexible package, you'd be better served with limma, DESeq2, or edgeR.

My interpretation of the cuffdiff manual says this is not true.

"If you have more than one replicate for a sample, supply the SAM files for the sample as a single comma-separated list."

-T/--time-series Instructs Cuffdiff to analyze the provided samples as a time series, rather than testing for differences between all pairs of samples. Samples should be provided in increasing time order at the command line (e.g first time point SAM, second timepoint SAM, etc.)

http://cufflinks.cbcb.umd.edu/manual.html

So my interpretation is that you could analyse a time-course with repeats for each time point, by supplying the individual BAM/CXB files for a time point as a comma-separated list, separating the time points by a space, and using the -T flag. Like so:

cuffdiff -T time1rep1.bam,time1rep2.bam,time1rep3.bam time2rep1.bam,time2rep2.bam,time2rep3.bam time3rep1.bam,time3rep2.bam,time3rep3.bam

Going to send this to [email protected] to get a confirmation of this as I haven't personally tested this and I'm not sure if my interpretation is correct.

[pocket_helix]

Any update from this query? I am having the same trouble coding a time series with multiple conditions and replicates.