[bioparser::FastqParser] error: too small chunk size!

[cbirbes]

Hi, I'm using Racon to polish bovine genome and i'm actually trying to find the "optimal" set of long reads to have the best time / quality ratio of polishing. For this i reduced my inital dataset by removing reads <10Kb and I encountered the following error : [bioparser::FastqParser] error: too small chunk size!

I tried to find where the problem could come from but nothing was really usefull.

Polishing with the initial dataset (with reads <10Kb) worked well. Does Racon need shorts reads to work ?

[rvaser]

Hello, Racon does not care about the read length. The error you encountered is in the parser. Can you tell me which version of Racon you are using?

Best regards, Robert

[cbirbes]

Thanks for your answer, i'm using the version 1.3.1

[rvaser]

Can you please update to the latest version? This should fix your problem.

[cbirbes]

i tried with the version 1.4.3 and still got the same problem. My command are there : module load bioinfo/racon-v1.4.3 racon -t 32 longreads.rename.fq.gz map.polisher.sam offspring_raw.cgt.fa > assembly.racon1.fa

And the command error : [racon::Polisher::initialize] loaded target sequences 15.813169 s terminate called after throwing an instance of 'std::invalid_argument' what(): [bioparser::FastqParser] error: too small chunk size! /work2/project/seqoccin/assemblies/polishing_1/bos_taurus/nfPolishing_trio1_mother_37165_Filtered_reads/work/75/2164e959c26c27ebfc3f0d01f1b58f/.command.sh : ligne 3 : 80759 Abandon

[rvaser]

The newest version (1.4.10) is at https://github.com/lbcb-sci/racon. Version 1.4.3 has the old parser with the bug :/

[cbirbes]

I did it with Version 1.4.10 but a new problem appear :/

[racon::Polisher::initialize] loaded target sequences 13.671844 s /work2/project/seqoccin/assemblies/polishing_1/bos_taurus/nfPolishing_trio1_mother_37165_Filtered_reads/work/c2/71cfaf5aebce87804d3c7850af1e02/.command.sh : ligne 3 : 93641 Erreur de segmentation racon -t 32 longreads.rename.fq.gz map.polisher.sam offspring_raw.cgt.fa > assembly.racon1.fa

This command worked well on other datasets with old version, do you know where the problem may come from ?

[rvaser]

Could be anything. How did you obtain map.polisher.sam? How large are your files? Did you modify you read file? The error you got before might indicate that one of your reads is too large to process.

[cbirbes]

Reads = ONT reads (36G) map.polisher.sam from minimap2 : minimap2 -a -t 16 -x map-ont offspring_raw.cgt.fa longreads.rename.fq.gz > map.polisher.sam (187G)

No modification on reads

[rvaser]

Please run zcat longreads.rename.fq.gz | head -n 12 and please check if the reads have 4 lines or more.

[cbirbes]

Ohhhh you're right. There is my fault. I did a python script to filter my fastq read and forgot to print as fastq file longreads.rename.fq.gz have a failed fasta format. Thanks for your help ! I come back to you if I have a new problem after checking whole of my files !

[rvaser]

No problem :)