search

BrooksLabUCSC / flair

Full-Length Alternative Isoform analysis of RNA
Other
205 stars 71 forks source link

diffSplice showing the same value for all replicates #235

Closed fburdet closed 1 year ago

fburdet commented 1 year ago

Hello,

I have an ONT experiment with 2 conditions, 4 replicates per conditions.

I used reads from pychopper, alidnted, corrected, merged (I used the 8 replicates to generate a bed file), collapsed into an isoforms bed, quantified.

Then I used diffSplice using the collapsed bed and the counts.tsv table, and the --test option.

Everything ran without an error.

What is weird is in the 4 drimseq files. The 4 replicates always have the same value, eg

==> drimseq_ir_DKO_v_WT.tsv <== feature_id gene_id DKO1_DKO_batch1 DKO2_DKO_batch1 DKO3_DKO_batch1 DKO4_DKO_batch1 WT1_WT_batch1 WT2_WT_batch1 WT3_WT_batch1 WT4_WT_batch1 lr adj_pvalue exclusion_chr14:47006354-47007289+ chr14:47006354-47007289 0.953 0.953 0.953 0.953 0.472 0.4720.472 0.472 506.01 1.02e-107 inclusion_chr14:47006354-47007289+ chr14:47006354-47007289 0.047 0.047 0.047 0.047 0.528 0.5280.528 0.528 506.01 1.02e-107 exclusion_chr14:47002497-47004632+ chr14:47002497-47004632 0.961 0.961 0.961 0.961 0.537 0.5370.537 0.537 451.98 2.93e-96 inclusion_chr14:47002497-47004632+ chr14:47002497-47004632 0.039 0.039 0.039 0.039 0.463 0.4630

That looks suspiciously weird... is that normal?

Thanks in advance!

Jeltje commented 1 year ago

It is normal behavior, see #234