Seqfilter is a small tool written in C on top of the excellent klib library by Heng Li. It allows filtering of fasta and fastq files based on sequence IDs and sequence length. The fasta and fastq input files may be gzipped.
git clone --recursive https://github.com/clwgg/seqfilter
cd seqfilter
makeSeqfilter will always try to filter by IDs, since thats its intended purpose. You can however omit the ID file. In that case, you have to specify the ‘-n’ flag for negative filtering if you want to produce output, to allow output of sequences without a “valid” ID.
Negative filtering (‘-n’) means that all sequences without ID matches are kept (subsequently, if no ID file is supplied, all sequences are without ID matches).
# filter by ID file
seqfilter -i in.fq -l ids.txt -o out.fq
#filter by ID file and min length 30
seqfilter -m 30 -i in.fq -l ids.txt -o out.fq
#filter only by min length 30
seqfilter -n -m 30 -i in.fq -o out.fq
#keep only sequence called "mt"
seqfilter -i in.fa -l <(printf "mt\n") -o mt.fa
#remove sequence called "mt" from input
seqfilter -n -i in.fa -l <(printf "mt\n") -o in_no-mt.fa