filter out FASTQ reads which are shorter

Question

I tried to filter out FASTQ reads which are shorter than 259 bp with

bioawk -cfastx 'length() >= 259 {print "@" " " "\n""\n+\n"}' good/SZ005_NoIndex_L002_R1_009.good.fq.gz \
  | gzip >  good-filtered/SZ005_NoIndex_L002_R1_009.good-filtered.fq.gz`. 

Unfortunately, the output was wrong:

@ 

+

@ 

+

@ 

What did I miss?

Thank you in advance,

Answer 1

If you have paired-end reads, this solution keeps the two files in-sync (i.e. discard pairs where one of the two reads is shorter than 259). Also, it uses only Unix tools without requiring external programs:

paste <(zcat S1_L001_R1_001.fastq.gz | paste - - - -) <(zcat S1_L001_R2_001.fastq.gz | paste - - - -) \
| awk -v FS='\t' 'length($2) >= 259 && length($6) >= 259' \
| tee >(cut -f 1-4 | tr '\t' '\n' | gzip > r1.fq.gz) \
| cut -f 5-8 | tr '\t' '\n' | gzip > r2.fq.gz

Answer 2

You need to use variable $seq. For example:

bioawk -cfastx 'length($seq)>=259{print "@"$name"\n"$seq"\n+\n"$qual}' test.fq.gz

There are also more convenient tools like seqtk and seqkit. With those, you may

seqtk seq -L260 test.fq.gz