trying to remove empty fasta sequence then creating a new file
I am working in python
from Bio import *
for current_seq in SeqIO.parse("mysequences.fasta", "fasta"):
if current_seq.seq.tostring() != '':
print( ">", current_seq.id, "\n", current_seq.seq.tostring())
I have this so far but there is an error. Also how would I create a file of the cleaned up version?
2 answers
outstide python:
$ cutadapt -m 1 test.fa
$ awk -v RS='>' '{print ($2!="")?">"$1"\n"$2:""}' test.fa | awk NF
$ seqkit -w 0 seq -gm 1 test.fa
>chr2
TAGCTAGCT
For python:
from Bio import SeqIO
with open("new.fa", "w") as o:
for current_seq in SeqIO.parse("test.fa", "fasta"):
if len(current_seq.seq) != 0 :
SeqIO.write(current_seq, o, "fasta")
test.fa is input fasta and new.fa is output fasta. Output fasta will be written in the same directory. If you are familiar with list comprehension in python, following is the way:
from Bio import SeqIO
old_seq=SeqIO.parse("test.fa", "fasta")
new_seq=[seq for seq in old_seq if len(seq.seq) != 0]
SeqIO.write(new_seq, "new.fa", "fasta")
Please read biopython documentation/tutorial. https://biopython.org/wiki/SeqIO. There is an example for filtering sequences by length.
Log in to answer this question.
More posts like this
-
finding overlapping motifs to increase length of motif
written by andrea •
-
'SeqRecord' object has no attribute 'transcribe'
written by ninideaninidea •
-
When running this code on a fasta file and a max() arg error occurs
written by rt2421 •
-
Error when converting/translating .tsv ORFs to .faa
written by anabaena •
-
blast using python
written by santhoshhegde278 •
-
trouble about biopython fasta file import
written by mendel18221884 •
-
one hot encoding for human genome
written by eneskemalergin •
-
trying to select some files from a multifasta depending on accession number using python
written by sumafa •
-
Seqio.Parse Some Error
written by vrutantshah •
-
Trouble Calling Emboss Program From Python
written by charles.bridges
Please clarify if you need assistance with python code or other solutions will work.
Use
101010button to formatcodeportion of your post properly. I have done it for you this time.First you want to collect the good records as a list and then write to file. Biopython writing FASTA files is quite well covered in the the documentation here. The first example there under 'Examples: Input/Output Example - Filtering by sequence length' is fairly similar to your needs and so it should be straightforward to adapt.