Tweet
I have 4 samples, each in triplicate (A1, A2, A3; B1, B2, B3; C1, C2, C3; D1, D2, D3)
When I use Cuffdiff to compare within each sample, say A1 to A2 or B2 to B3, I get DE genes and most of the 30K or so genes have FPKM values. This is true for all 12 .bam files, in other words there is a similar result (most genes have FPKM values and status of OK) for all in-sample comparisons.
When I use Cuffdiff to compare between samples, say A1,A2 to C2,C3 or A1,A2,A3 to D1,D2,D3, I get results only for A1,A2 to B2,B3, but for A2,A3 to B2,B3, the status for almost all genes is NOTEST.
The only genes with status OK and FPKM values are on mitochondria and chloroplast DNA. This same result occurs for all other between sample comparisons. (Only A1,A2 to B2,B3 works).
The same merged.gtf and genome.fa files are being used in all Cuffdiff's. Only the accepted_hits.bam files are changing.
When I use Cuffdiff to compare within each sample, say A1 to A2 or B2 to B3, I get DE genes and most of the 30K or so genes have FPKM values. This is true for all 12 .bam files, in other words there is a similar result (most genes have FPKM values and status of OK) for all in-sample comparisons.
When I use Cuffdiff to compare between samples, say A1,A2 to C2,C3 or A1,A2,A3 to D1,D2,D3, I get results only for A1,A2 to B2,B3, but for A2,A3 to B2,B3, the status for almost all genes is NOTEST.
The only genes with status OK and FPKM values are on mitochondria and chloroplast DNA. This same result occurs for all other between sample comparisons. (Only A1,A2 to B2,B3 works).
The same merged.gtf and genome.fa files are being used in all Cuffdiff's. Only the accepted_hits.bam files are changing.