search

jhkorhonen / MOODS

MOODS: Motif Occurrence Detection Suite
Other
95 stars 12 forks source link

moods-dna.py: Segmentation fault (core dumped) for long motifs #45

Open osyafinkelberg opened 1 year ago

osyafinkelberg commented 1 year ago

Hello!

I am scanning the human genome for the long CTCF motif (34 bp) using a PWM-format matrix:

moods-dna.py --sep ";" -s hg38.fna --batch --bg 0.25 0.25 0.25 0.25 --p-value 0.0001 -S MA1929.1_pwm -o ctcf_scan

It gives the following error: Segmentation fault (core dumped).

Presumably that is because of the motif length. If so, what is the maximum motif length accepted by the moods-dna.py ? I also wanted to ask if there is a way I can use moods-dna.py (or other MOODs soft) to search for a motif with a non-informative spacer? For example motif MA1929.1 contains a 6-7 bp spacer, so that effectively I could search for 27-28 bp motif instead of a 34-bp one.

Thank you very much!

jhkorhonen commented 1 year ago

There isn't hard limit on the motif size, though at some point much of the algorithm engineering in MOODS becomes rather redundant. I verified that I could indeed run MOODS with the motif in question with no issues, so something else is going on here.

Could you:

  1. Verify that you can run MOODS successfully with other inputs, e.g. using the example inputs?
  2. Provide your actual input file MA1929.1_pwm for reference?
  3. Provide full logs given by MOODS for a run where the segmentation fault occurs (use parameter -vvv)?
osyafinkelberg commented 1 year ago

Oh, I found out that I used the incorrect path for the PWM-file... Sorry to bother you!