Entering edit mode
3.9 years ago
demoraesdiogo2017
▴
110
Hello
Two of my samples had very low alignment percentage (44 and 32%). I asked HISAT2 to make separate files for unaligned reads which are on fastqsanger formats. I would like to BLAST these unaligned reads to have an idea where this contamination came from. I would like to use NCBI BLAST+ blastn on usegalaxy to do so. However, the unaligned reads (in fastqsanger format) that HISAT2 apparently are not suitable for this. I tried uncompressing these files, but still can't use them. Any suggestions?
You could try converting them to fasta, e.g. with the
seqtk_seq
tool.Generally there is no need to blast every unaligned read. If you suspect there is contamination you could take a sample of reads (~10) and use web blast at NCBI. You should be able to view your data in galaxy and grab a set of reads. You can easily convert them to fasta format by changing
@
beginning of fastq header to>
and then getting rid of the+
and the q-score string in lines 3 and 4 from fastq records.