TimoLassmann/kalign

Segmentation fault for large datasets

Closed this issue · 7 comments

Hey,
I ran multiple sequence alignments on protein sequences of length 7096 for largedatasets (1000's, 100,000's) on clustalO and kalign. I had cleaned the datasets i.e. removed sequences having BJOUXZ characters and duplicate sequences. Running MSA for 15,000 sequences and above gave me segmentation fault for building guide tree on kalign whereas, clustalO ran just fine.
[2022-03-10 11:55:18] : LOG : Building guide tree.
Segmentation fault

Could you share the sequences with me so I can reproduce the problem? Thanks

Thanks! I think I identified the problem; working on a solution. This may take a few days.

Dear Shubhangi Kandwal,
I fixed the problem in the latest release (3.3.2). Let me know if this works on your end.
Thanks for bringing this to me attention!
T

Great. Any other issues let me know!