Renaming Entries In A Fasta File
5
10
Entering edit mode
8.9 years ago
thiago84naka ▴ 110

Hello,

I have never made a script in my life.

The ploblem is how to change the fasta names like this input file:

>Glyma04g14800|Glyma04g14800.3
MMLETVAAVPGMVAGMLLHCKSLRRFEHSGGWIKALLEEAENERMHLMTFMEVAKPKWYE
>Glyma05g24460|Glyma05g24460.1
SNVSIDLTKHHVPKNFLDKVAYRTVKLLRIPTDLFFKRRYGCRAMMLETVAAVPGMVGGM


in this output file (change original names to numbers in ascending order, starting with 1):

>1
MMLETVAAVPGMVAGMLLHCKSLRRFEHSGGWIKALLEEAENERMHLMTFMEVAKPKWYE
>2
SNVSIDLTKHHVPKNFLDKVAYRTVKLLRIPTDLFFKRRYGCRAMMLETVAAVPGMVGGM


I'm so grateful for helping. Regards, Naka

fasta • 44k views
33
Entering edit mode
8.9 years ago
awk '/^>/{print ">" ++i; next}{print}' < file.fasta

0
Entering edit mode

Sir can we modify above awk syntax in this way

>1
>2
....


it prints like

>chromosome1
>chromosome2
...


for that purpose, where and how do I put the text "chromosome"

2
Entering edit mode

@Raghav: If you wanted to add chromosome in the header with the counter, simply add it in the ">" portion of the one-liner.

awk '/^>/{print ">chromosome" ++i; next}{print}' < file.fasta

0
Entering edit mode

How can we add "chr" just after >? I don't want to change anything else. For example:

2L I want it to become chr2L

0
Entering edit mode

Hello there, (I already solved this)

I am trying to understand your script line to modified. So, is the script saying?:

For every line ('/) where you find a > (^>/) print the > and then add (+) a counter (+), then next print what follows.

In my case the names are like:

> M02137:143:000000000-APU54:1:1101:21985:13014 1:N:0:10
> M02137:143:000000000-APU54:1:1112:18691:9995 1:N:0:10


etc.

I want to leave only what is different.

awk '/^>/{print ">" remove "M02137:143:000000000-APU54:1:"; next}{print}' < file.fasta


And can I do this in ssh? ( I don't think I have awk installed)

Caro

PS: I am new to HTS/NGS and don't know much about programming

0
Entering edit mode

This doesn't work if the read spans in multiple lines ?

0
Entering edit mode

@Pierre Lindenbaum

Hi,

1. how can I modify this command to add genus_species name after > in every entry and yet keep most of the information in the the header

ie. my entries are like this

>lcl|HF546977.1_cds_CCO27433.1_1 [gene=cox1] [protein=cytochrome c oxidase subunit 1] [protein_id=CCO27433.1] [location.......]


and want to have the entries name like this

>genus_species HF546977.1_cds_CCO27433.1_1 [gene=cox1] [protein=cytochrome c oxidase subunit 1]


By using

awk '/^>/{print ">genus_species gene." ++i; next}{print}' < file.fa


I got,

>genus_species gene.1


and so on

2. and how can I add output file in the command line

Having the genus_species name in the beginning is requires as I'll be comparing different species and also, I don't want to loose the ids and protein names for ease of downstream analysis.

0
Entering edit mode

Hello Pierre, Thank you for your useful code. May I please ask how can I modify the code to keep everything else in the sequence and just to add the sample name in front and that too for the batch of files.

e.g. my file looks like :

>M03691:51:000000000-BD94Y:1:1101:14841:1381 1:N:0:1


ACTGGGTGTAAAGGGCGTGTAGGCGGAGAAGCAAGTCAGAAGTGAAATCCATGGGCTTAACCCATGAACTGCTTTTGAAACTGTTTCCCTTGAGTATCGGAGAGGCAGGCGGAATTCCTAGTGTAGCGGTGAAATGCGTAGATATTAGGAGGAACACCAGTGGCGAAGGCGGCCTGCTGGACGACAACTGACGCTGAGGCGCGAAAGCGTGGGGAGCAAACAGGATTAGATACCCCGGT

M03691:51:000000000-BD94Y:1:1101:15960:1389 1:N:0:1 TACTGGGGTATCTAATCCTATTTGCTCCCCACGCTTTCGGGACTGAGCGTCAGTTATGCGCCAGATCGTCGCCTTCGCCACTGGTGTTCCTCCATATATCTACGCATTTCACCGCTACACATGGAATTCCACGATCCTCTCACACACTCTAGCTCTACGGTTTCCATGGCTTACCGAAGTTAAGCTTCGATCTTTCACCACAGACCCTTAGTGCCGCCTGCTCCCTCTTTACACCCAGT M03691:51:000000000-BD94Y:1:1101:15662:1415 1:N:0:1 ACTGGGTGTAAAGGGCTCGTAGGCGGTTCGTCGCGTCCGGTGTGAAAGTCCATCGCTTAACGGTGGATCTGCGCCGGGTACGGGCGGGCTGGAGTGCGGTAGGGGAGACTGGAATTCCCGGTGTAACGGTGGAATGTGTAGATATCGGGAAGAACACCAATGGCGAAGGCAGGTCTCTGGGCCGTTACTGACGCTGAGGAGCGAAAGCGTGGGGAGCGAACAGGATTAGATACCCCCGTA

Now I want to add Sample name after > and keep everything else as it it.

This process I want to do for a batch of files. Any help will be really great. Thanks, Mitra

0
Entering edit mode

Hello Pierre, when I use awk '/^>/{print ">" ++i; next}{print}' < file.fasta, the changes are made but not saved. I want to distinguish between two numbered contig.fasta files (each fasta is numbered 'contig 00001', 'contig00002' etc, I want to name the 1st contig.fasta 'shorter_contig 00001, 'shorter_contig 00002' and the 2nd.contig.fasta 'longer_contig 00001, 'longer_contig 00002' ) is there a way make the header modifications permanent? Thanks

0
Entering edit mode

the changes are made but not saved.

http://wiki.bash-hackers.org/howto/redirection_tutorial

0
Entering edit mode

Hello, i want to change the fasta name of this input file :

M04631:312:000000000-C6V6K:1:2107:11495:1734 1:N:0:ACTGAGCG+TTATGCGA

M04631:312:000000000-C6V6K:1:2107:13059:1785 1:N:0:ACTGAGCG+TTATGCGA

In this fasta name:

what script do I have to use??

1
Entering edit mode

With respect, if you are already stuck at this most simple task, better spend some quality time on Unix and NGS basics before diving into any analysis. In the end, you as the analyst have to stand up for your analysis.

0
Entering edit mode

A similar question+answer was posted above. What are you missing ?

0
Entering edit mode

my question was resolved, thank you. I needed it in this manner because i have other script that only works with these fasta names . thanks.

8
Entering edit mode
8.9 years ago

The Fastx Renamer tool can do this as well: http://hannonlab.cshl.edu/fastx_toolkit/commandline.html#fastx_renamer_usage

$more test.fa >GS6SIDE04J1T1R xy=4004_1485 CATAGTAGTGAGAGTTGATCATGGCTCAGCCATCTCATCCAGCAGCCGCGGTAATCACTACTAT >GS6SIDE04J0352 xy=3996_712 ACGAGTGCGTAGAGTTGATCATGGCTCAGCAGCCTCCTCGTGCCAGCAGCCGCGGTAATACGCACTCG >GS6SIDE04JM7EM xy=3837_2988 AGCACTGTAGAGAGTTGATCCTGGCTCAGGGATAGGCCAGCAGCCGCGGTAATCTACAGTGC$ ~/Downloads/bin/fastx_renamer -i test.fa -n COUNT
>1
CATAGTAGTGAGAGTTGATCATGGCTCAGCCATCTCATCCAGCAGCCGCGGTAATCACTACTAT
>2
ACGAGTGCGTAGAGTTGATCATGGCTCAGCAGCCTCCTCGTGCCAGCAGCCGCGGTAATACGCACTCG
>3
AGCACTGTAGAGAGTTGATCCTGGCTCAGGGATAGGCCAGCAGCCGCGGTAATCTACAGTGC

6
Entering edit mode
8.9 years ago

Try this:

cat youFile.fa | perl -ane 'if(/\>/){$a++;print ">$a\n"}else{print;}' > youFile_new.fa

8
Entering edit mode

Instead of useless cat, try: perl -ane 'if(/\>/){$a++;print ">$a\n"}else{print;}' youFile.fa > youFile_new.fa

5
Entering edit mode
5.3 years ago

Here is a generic way to convert ncbi headers to simple header

>gi|1002620271|ref|NC_029525.1| Coturnix japonica isolate 7356 chromosome 10, Coturnix japonica 2.0, whole genome shotgun sequence
TACTCCCCAAGAA


to

>NC_029525.1
TACTCCCCAAGAA


By sed :

sed 's/^[^ ]*[|]$$[^|]*$$[|] .*\$/>\1/' Coturnix_japonica.fasta > Coturnix_japonica_rename.fasta

0
Entering edit mode

@noirot.celine Hi, I have a similar problem and your above command didn't work for me (I am really new to linux environment). I have different fasta files. Some of my fasta headers are like this (augustus output file)

g1134t1 geneg1134 I want to keep the header and just add the species_genus name after >

or better like this

Species_genus gene1134

Similarly, for file with headers like this,

AG1IA_00006 contig1:1338:4722:+ [translate_table: standard]

I want to keep >AG1IA_00006 and since the ids in files are also not in continuation, so simply renaming in series won't help.

p.s. my OS= Ubuntu16.04

1
Entering edit mode
8.9 years ago
AGS ▴ 230

I'd use faSimplify