OK, I will definitely try. Thanks a lot !!!!!
• 0 views
•
link
written
by
arraychip •
split fasta of a protein sequence into consecutive n amino acids
let's say I have a fasta of a protein sequence
> albumin
MKWVTFISLL FLFSSAYSRG ... ... ...
I want to split the sequence into all possible consecutive 8 amino acids (only in 1 direction, amino -> carboxyl) (And no looping(I don't know if it is the right expression), say, GMKWVTFIS)
I need
> fasta.albumin1
MKWVTFIS
>fasta.albumin2
KWVTFISL
> fasta.albumin3
WVTFISLLF
...
> fasta.albumin13
FSSAYSRG
And, I want to do this for all known human protein sequences.
How would I do it??? I need the result as a fasta or fasta files. And IDs for resulting 8-mer seuqeunces need to be unique.
• 5,248 views
•
link
updated
by
Pierre Lindenbaum
•
written
by
arraychip •
•
written
by
arraychip •
3 answers
There are many methods. For long terms, some programming skill is worth learning.
Here's a simple way by using seqkit.
$ seqkit sliding -s 1 -W 8 seqs.fa
>albumin_sliding:1-8
MKWVTFIS
>albumin_sliding:2-9
KWVTFISL
...
>albumin_sliding:13-20
FSSAYSRG
• 0 views
•
link
written
by
shenwei356
You requirement is to generate simple kmers from sequences. BioPython solution. Please change last print statement to: print('>'+strrecord.id)+'|kmer_'+str(count)+'\n'+str(my_kmer))
#!/usr/bin/env python3
from Bio import SeqIO
myfile=SeqIO.parse('test.fa','fasta')
for record in myfile:
sequence=record.seq
seq_len=len(sequence)
#define the kmer len
kmer=8
count=0
for seq in list(range(seq_len-(kmer-1))):
count=count+1
my_kmer=(sequence[seq:seq+kmer])
print('>'+strrecord.id)+'|kmer_'+str(count)+'\n'+str(my_kmer))
I know a little bit of R. But maybe this is a good chance to study (Bio)Python. I will definitely try, once I study a little bit of BioPython. Thanks a lot
• 0 views
•
link
written
by
arraychip •
Using bioalcidaejdk:
$ java -jar dist/bioalcidaejdk.jar -e 'final int N=8;stream().filter(F->F.length()>=N).forEach(F->IntStream.range(0,F.length()-N).forEach(L->{out.println(">"+F.getName()+"."+(L+1)+"\n"+F.subSequence(L,L+N));}));' input.fasta
• 0 views
•
link
written
by
Pierre Lindenbaum
Log in to answer this question.
More posts like this
-
Am I understanding PCA for protein analysis correctly?
written by Sam •
-
Degenerate/combinatorial protein BLAST search
written by Albert •
-
Compare protein functionality between species
written by Howard •
-
Extracting amino acid positions from a fasta sequence
written by akang
-
Protein to RNA codons
written by shiningsky000 •
-
Codon usage bias: Am I seeing decreased translation due to my protein soaking up tRNAs?
written by sako242 •
-
Bi-gram feature extraction from a protein sequence.
written by arijeetsat •
-
blastp word_size parameter seems to be ignored - BLASTP 2.4.0+
written by matthys.potgieter •
-
How I can make an open reading frame finder with Python ?
written by Kevin_Smith •
-
Eukaryotic Cell From Scratch
written by caters123 •
If you know python/perl then this should be easy to do. Someone is bound to put a program up as answer shortly. @Pierre will likely have a one liner in
awk.Have you tried to do this yourself? Please post your code you have already tried.
I am a biologist. I can google to search for news, but can't, and don't know any scripting. I was even struggling with my posting questions. Sorry
That is fine. This forum is there to help with that. I have formatted your post yet again. Please do not change the formatting back to plain text.
You say
5AA in title but8in the body of post. Which is it?There are ~20K+ validated human proteins at UniProt so results file is going to be gigantic.
Note: Why do you keep changing back the formatting of the data in your post to plain text?
Yes, it's going to be a 'big data' I really want try.
Hey @genomax
There is some issue with the code formatter below. A '(' is missing in my code
print('>'+strrecord.id)+'|kmer_'+str(count)+'\n'+str(my_kmer))below but looks okay when I check in the edit mode.There is a known issue with python code formatting and biostars display of that code. I think the fix is to put a comment there to add the
(at proper place. Not elegant but will work for now.Edit: Best solution is to put your code in a gist and then link that in your post. Biostars then looks the code up via
gistand formats it correctly.I see! I will opt for the
gistoption next time. Thanks