Remove sequences <300 bases from FASTA file

Question

[zoppisemma]

I have a multiple FASTA file containing contigs deriving from metagenomic data. I need to remove all contigs less than 300 bp long. Ho do I proceed?

[lakhujanivijay]

Hi zoppisemma

There are multiple solutions provided by different users. you should upvote/ accept answers which helped. This will help others looking for such solutions.

[Jeffin Rockey]

See this post and tweak for 300.

[Ram]

This should be just a comment and not an answer, as you're only pointing to an existing post/answer. I've moved it to one.

[Jeffin Rockey]

Thanks for the correction Ram.

[Pierre Lindenbaum]

duplicate: How To Filter Multi Fasta By Length??

[Hugo]

Other people gave you excellent solutions. Nevertheless, you may be also interested in SEDA (http://www.sing-group.org/seda/ ), an open-source tool for processing FASTA files. Among other functions, it provides an operation to apply different filters, including sequence length (https://www.sing-group.org/seda/manual/operations.html#filtering ). Regards.

Answer 1

[lakhujanivijay]

using seqkit

seqkit seq -m 300 your_fasta.fa

download here

Answer 2

[GenoMax]

Using reformat.sh from BBMap suite.

reformat.sh in=your.fa out=filtered.fa minlength=300

Answer 3

[harish]

Hi!

You can use seqtk for the same. The command should be:

seqtk seq -L 300 contigs.fasta > file.fasta

[lakhujanivijay]

@harish

Just FYI (for larger datasets), see this (seqkit benchmark)

https://bioinf.shenwei.me/seqkit/#benchmark

[harish]

Ahh. That's nice. Glad to learn something new today!

[lakhujanivijay]

I am using it for all sorts of fasta/q manipulation and found it really fast and effective.

Answer 4

[finswimmer]

awk solution which should work for multiline fasta files:

awk -v RS=">" -v FS="\n" '{for(i=2;i<NF;i++) {l+=length($i)}; if(l>300) printf ">%s", $0}' test.fasta