How to extract out gene specific reads from concatenated fastq file

Question

[Mo]

Hi all,

I have expressed a library of a heterologous gene in the HEK293 cell line and performed direct cDNA sequencing using a nanopore direct cDNA sequencing (DCS109) kit. I wanted to know what people do to extract their gene-specific reads from concatenated fastq files.

I am using the blast method but it's giving me some issues, I wanted to know what are the other options.

Apologies for this basic question, your suggestion will be really helpful.

Thanks,

Answer 1

[GenoMax]

With nanopore reads you should use minimap2 (LINK) which is the best aligner for nanopore data. Write out the results in SAM format and then extract reads that align to your gene of interest.

[Mo]

Hi Max, Thanks for your reply. I also use minimap2 but before that, I blast the common region from my gene of interest against all reads in the fastq file. I do this to determine whether the strand sequenced is a forward or reverse strand. Then I convert all the reverse reads to forward reads and concatenate them with the rest of the forward reads. After that, I do minimap alignments. The strand determination and conversion of all reads to forward reads helps me to align them with the gene of interest with barcodes.

Do you think this is also possible with minimap to determine the strand direction? Thanks a lot.

Answer 2

[swbarnes2]

People generally don't parse the raw fastq directly like that. Use some kind of aligner to get assignments for all of the reads in the fastq, then you can pull out what you want.