search

bioinfologics / w2rap-contigger

An Illumina PE genome contig assembler, can handle large (17Gbp) complex (hexaploid) genomes.
http://bioinfologics.github.io/the-w2rap-contigger/
MIT License
44 stars 14 forks source link

Segfault during Step 6: removing small components #43

Closed melop closed 4 years ago

melop commented 4 years ago

Hello,

w2rap-contigger (downloaded 2019-10-24) ran Segfault during step 6 on a small dataset. I am attaching the log file and the fastq files that caused the error. assembly.log R1.fq.gz R2.fq.gz

Best Regards, R. Cui

jonwright99 commented 4 years ago

Hi, Can you give a bit more information about what you are trying to do? Your read files are pretty small so are you trying to assemble a short section of sequence or are these reads a subset of a larger dataset? Thanks, Jon

melop commented 4 years ago

Dear Jon,

Thank you for your reply. I was trying to extract reads from targeted regions of the genome, and perform a local assembly. w2rap-contigger worked fine on 6 other similar datasets, but only failed on this one.

Best, R. Cui

melop commented 4 years ago

Dear Jon,

I have tested now using another two independent datasets from the same target region from the same species (but different individual). The program segfault at the same step. I am attaching these two additional datasets for you to look at. I notice that by setting --min_freq to 4 the program can finish, but only very short contigs were returned and N50 is "0". debug.tar.gz

Best, Ray

jonwright99 commented 4 years ago

Hi Ray, Have you tried using a different assembler on these datasets, especially the ones that fail. It could be that this region is just not able to be assembled with the data you have. W2rap is designed to efficiently assemble large, complex genomes so probably isn't the tool for small local assemblies, I would suggest something much simpler like velvet would do a better job. Best, Jon

melop commented 4 years ago

Hi Jon,

I tested 6 such datasets from other species, including species that are closely related, and they all worked with w2rap-contigger. I will also try velvet to see if they assemble. Thanks, Ray

melop commented 4 years ago

Hi Jon, Velvet was able to run through without errors. But yeah the contigs tend to be very short as expected. Does that explain the segfault of w2rap-contigger perhaps?

Ray

jonwright99 commented 4 years ago

Great! The segfault you are seeing in the contigger is because no contigs are being generated at an earlier stage. This isn't ideal and we've fixed the issue in our current development branch. We're planning to make the development branch the master branch once we've finished up our latest changes and fixes, hopefully soon.