The four DE output files isoform_, gene_, tss_groupd_ and cds_exp all address different levels of pooling matches between RNAseq reads and genome.
I have genes of known DE (QPCR) scattered in all 4 .diff files. To make a list of ALL putative DE genes identified by cuffdiff should I blend the results from all 4 .diff files? Here is my post-cuffdiff workflow:
1. Copy 4 .diff files into excel tabs. The non_exp .diff files have nothing significant.
2. filter for significance
3. copy all significant rows into a new tab.
4. copy all significant rows into one combined tab.
5. sort out duplicates from combined list
6. filter on fold change to rank.
This fails to calculate a mean of numeric values and is a pretty round-a-bout way to get a simple fold change list.
I have genes of known DE (QPCR) scattered in all 4 .diff files. To make a list of ALL putative DE genes identified by cuffdiff should I blend the results from all 4 .diff files? Here is my post-cuffdiff workflow:
1. Copy 4 .diff files into excel tabs. The non_exp .diff files have nothing significant.
2. filter for significance
3. copy all significant rows into a new tab.
4. copy all significant rows into one combined tab.
5. sort out duplicates from combined list
6. filter on fold change to rank.
This fails to calculate a mean of numeric values and is a pretty round-a-bout way to get a simple fold change list.