kvittingseerup / IsoformSwitchAnalyzeR

An R package to Identify, Annoatate and Visialize Isoform Switches with Functional Consequences (from RNA-seq data)
100 stars 18 forks source link

The isoform_id in isoformFeatures and exons does not match #239

Open mpage21 opened 4 months ago

mpage21 commented 4 months ago

@kvittingseerup We are trying to run this tool with output from Bambu but we are getting this error:

Error in createSwitchAnalyzeRlist(isoformFeatures = myIsoAnot, exons = myExons, :
The isoform_id in isoformFeatures and exons does not match

In our case, we have two types of IDs, ENST and Bambu ids. When we filter to just have the ENST ids we get past this step. When we filter to just have Bambu ids, we get past this step. But if we try to have both together, we get stuck here.

Do you have any inkling as to why this would be the case? Here is the command we ran:

switchList <- importRdata(
    countsMatrix,
    cpmMatrix, 
    designMatrix,
    "path/to/gtf/Homo_sapiens_plus_new_high_cofidence.GRCh38.109_EXON_CDS.gtf",                           
    ignoreAfterPeriod = TRUE)
chunxubioinfor commented 4 months ago

Hi Madeline, this error indicates that the isoform_id from the gtf file doesn't match the isoform_id from countsMatrix. Could you check the isoform_id in the gtf? Also, the argument ignoreAfterPeriod = TRUE should be careful while using. For further info, please check ?importRdata. If you have further questions, maybe you could share the files or the part of files then I can take a look.😊

Sparkle-27 commented 3 months ago

Hi @chunxubioinfor I have checked the isoform_id from the gtf and isoform_id in salmon$counts, however, there is no difference between them. When I only use the isoform from iso-seq data instead of including the ENS* transcript, everything runs correctly. Do you have any suggestion? Best image