I have a file with more than 15000nt sequence and i want it to be separated into 1000nt new files like F1,F2 .....
Example
MainFile
ATGCATATGCCCAGTAGCGAATGATGATCA ATGCATA_TCCCAGTAGTGAATGATAATCA _CAT_TGCC_ATTAGAGAATGATGATCA_C
INTO 3 FILES OF 10nt each
FILE1
ATGCATATGC ATGCATA_TC _CAT_TGCCAFile2
CCAGTAGCGA CCAGTAGCAG AGAGAATGAGFile3
ATGATGATCA ATGATAATCA GATGATCA_C
3 answers
- For the format you paste (one line for a sequence ), use the commands below (change
10to1000, change30to15000, change9to999). - For the FASTA format (usually produced by multiple sequence alignment software), change the
awkcommand with other tools, likeseqkit subseq -r $s:$e msa.fasta > msa.$s-$e.txt.
Just use awk:
for s in $(seq 1 10 30); do \
e=$(expr $s + 9); \
echo $s $e; \
done
1 10
11 20
21 30
for s in $(seq 1 10 30); do \
e=$(expr $s + 9); \
awk -v s=$s '{print substr($0, s, 10)}' msa.txt > msa.$s-$e.txt
done
more msa.*.txt
::::::::::::::
msa.1-10.txt
::::::::::::::
ATGCATATGC
ATGCATA_TC
_CAT_TGCC_
::::::::::::::
msa.11-20.txt
::::::::::::::
CCAGTAGCGA
CCAGTAGTGA
ATTAGAGAAT
::::::::::::::
msa.21-30.txt
::::::::::::::
ATGATGATCA
ATGATAATCA
GATGATCA_C
Thanks, but seqkit split is for splitting many sequences into several parts, rather than splitting sequences into fragments which are job of seqkit subseq or seqkit sliding.
input: test.fa
$ seq 1 10 30 | while read line; do (echo ">seq_"$line"_"`expr $line + 9` && cut -c $line-`expr $line + 9` test.fa | sed '/^$/d') > "seq_"$line"_"`expr $line + 9`".fa" ; done
$ tail -n+1 test.fa seq_*.fa
==> test.fa <==
>seq
ATGCATATGCCCAGTAGCGAATGATGATCA
ATGCATA_TCCCAGTAGTGAATGATAATCA
_CAT_TGCC_ATTAGAGAATGATGATCA_
==> seq_1_10.fa <==
>seq_1_10
>seq
ATGCATATGC
ATGCATA_TC
_CAT_TGCC_
==> seq_11_20.fa <==
>seq_11_20
CCAGTAGCGA
CCAGTAGTGA
ATTAGAGAAT
==> seq_21_30.fa <==
>seq_21_30
ATGATGATCA
ATGATAATCA
GATGATCA_
Remove original header from very first split fasta file (>seq here)
$ seq 1 10 30 | while read line; do (echo ">seq_"$line"_"`expr $line + 9` && cut -c $line-`expr $line + 9` test.fa | sed '/^$/d') > "seq_"$line"_"`expr $line + 9`".fa" ; done
- seq 1 10 30 - prints numbers from 1 to 30 with a window of 10 without overlaps
- bash loop has following logic: a) Echo the new file name with "seq", "number" from step 1 and "a second number generated by adding 9 to number from step1'. b) use cut to extract characters column wise from test.fa. Ranges are provided by number from step and another number generated by adding 9 to number from step 1 c) using sed remove the empty lines d) > outputs the file to a file named "seq" with number from step 1 and a second number generated by adding 9 to the number from step 1.
Logic of adding 9 to step 1 number is window size is 10 and step 1 numbers are 1, 11,21 and you would require 1-10, 11-20, 21-30 for cutting characters.
Thank you for your valuable time and explanation. Can you tell me how can i get the output in a format like seq0001 seq 00002 ...... and so on or seq_000001_001000.fas. Because when i run sort command in these file names. It sorts based on First character like
seq_110001_111000.fas
seq_11001_12000.fas.
seq_111001_112000.fas
and so on.
So its difficult to sort based on the file no.
man sort
-V, --version-sort
natural sort of (version) numbers within text
PyFasta might help: https://pypi.org/project/pyfasta/#command-line-interface
The main file is not a single line fasta file its a multiple sequence aligned file . i WANT TO CUT THEM BASED ON COLUMN i.e. horizontally 1000 chatacters each.
Example
ATGCATATGCCCAGTAGCGAATGATGATCA
ATGCATA_TCCCAGTAGTGAATGATAATCA
_CAT_TGCC_ATTAGAGAATGATGATCA_
-MainFile
INTO 3 FILES OF 10nt each
ATGCATATGC-FILE1
ATGCATA_TC
_CAT_TGCCA
CCAGTAGCGA-File2
CCAGTAGCAG
AGAGAATGAG
ATGATGATCA-File3
ATGATAATCA
GATGATCA_C
Log in to answer this question.