[Date Prev][Date Next] [Thread Prev][Thread Next] [Date Index] [Thread Index]

CI for fastx-toolkit [Outreachy]



Hello Andreas, 

I added test suite for fastx-toolkit package. But I left test for fastq_quality_filter commented for now, because it doesn't work.

When I run 

fastq_quality_filter -i galaxy/test-data/fastq_qual_filter1.fastq -q 33 -p 100

I've got: 

fastq_quality_filter: bug: got empty array at fastq_quality_filter.c:97


This happens, because the program works on assumption that value of quality lays between -50 and 50 (MIN_QUALITY_VALUE, MAX_QUALITY_VALUE). Meanwhile the function convert_ascii_quality_score_line computes: 

quality = (int) (ascii_quality_score - 33)

So for reads Solexa, Illumina 1.3-1.8 (Phred+64) computed quality could be more than 50. 

I guess the program just not compatible with Phred+64 format, though test-data folder contains a file in such format. I'm not sure If I should patch this with just increasing MAX_QUALITY_VALUE, or somehow to find a way to handle files in this format differently and compute quality for them as ascii_quality_score - 64. Or should I leave this decision to author of program and just specify somewhere for package that it works only with fastq files in Phred+33 format?

Regards, Nadiya Sitdykova






Reply to: