search

williamritchie / IRFinder

Detecting intron retention from RNA-Seq experiments
53 stars 25 forks source link

irfinder.stderr #56

Closed zhengjiantao closed 5 years ago

zhengjiantao commented 5 years ago

I got some errors in irfinder.stderr. Input error at line:1 Characters read on last read call:0

And I had all 0 in irfinder.stdout. such as image

My data is single-end reads RNA-Seq data. My code is as following IRFinder -m BuildRefProcess -r IRFinder/REF IRFinder -m BAM -r IRFinder/REF -d IRFinder/xx xx.sortedByName.bam What am I doing wrong?

zhengjiantao commented 5 years ago

@dg520

dg520 commented 5 years ago

@zhengjiantao ,

First thing to check, is your BAM file recording chromosome as chr1, chr2, .... while your GTF file used to build genome reference is recoding chromosome as 1, 2, ....? Or vise versa? They have to be in the same format.

If that's not the case, could you please provide me more information of your problem? Otherwise I can only guess what happened...Included but not limited to: You system setup:

  1. What's the version of your GCC?
  2. What's the version of your Bedtools? and STAR?

Your genome:

  1. what species are your using here? Is it a well-annotated genome?
  2. How many chromsomes and contigs in the GTF file?

Your RNASeq:

  1. What protocol was used to generate RNASeq reads? Poly-A enriched or Ribo depletion?
  2. What pipeline you've used to align your RNASeq reads? STAR or something else?
  3. How many uniquely aligned reads you ended up with?

About IRFinder:

  1. Did you encounter any problem when building the IRFinder reference. Is any file at IRFinder/REF/IRFinder empty?
  2. Did you encounter any problem when you run the quantification step of IRFinder?

Best, Dadi

zhengjiantao commented 5 years ago

@dg520 Thank you very much for your careful recovery. I have checked all the items listed above. Including some of the details of STAR. Ironically, the problem comes from parallelism in linux. I solved this using the "wait" command

Thanks again for your reply!