I have a huge FASTA file (190000 sequences), but the sequences have wraps in the middle. For that reason, the software I am using to analyze my NGS data (mothur) doesnt even recognize the file as fasta. Do you know how to fix the file? I am not a bioinformatic, so I need something relatively easy. Thank you!!!!
Sure, could you write or upload any example? I don't know exactly what you mean by "wraps"
Hello Sarah,
I assume you are asking how to transform an multiple line .fasta to a one line .fasta. There are several programs (e.g. Mesquite, MEGA, etc.) that can do this but since you are dealing with a large data set I assume you are doing everything from the command line. There are several simple ways to do this, but since I do not know your programming background (e.g. experience with awk, bash, perl, and/or python) I have place a link below that includes several different way to accomplish this task.
https://www.biostars.org/p/9262/
Hope this helps.
Thank you Max R. Bangs! I have no experience with programming..but I guess this is going to start changing! :P I will have a look to this link!
Galo!
What I mean is that the fasta sequence is not in one line. For example (I totally invented this sequences):
>Example_problematic_Fasta
AAATCATCGGG
AAAAATTTTTCCCCGGGGGG
>Example_Good_Fasta
AAATCATCGGGAAAAATTTTTCCCCGGGGGG
It is easy to fix when I have less that 1000 sequences...But with 190000 sequences it is not so easy...well, I guess it is easy if you know programming and you have LInux in the lab! :)
from programming point of view it is trivial, start learn any programming language (I recommend python for beginners, in this case after splitting multiple fasta into singular sequences it will be as easy as: "seq = seq.replace('\n', '')"
I guess you would like to have web based application, but the problem is that then you would need to send your sequences to server and back and if you have GB of data this will not work too well
see also https://www.biostars.org/p/9262/
Hi Sarah,
Attached is the python script to fix your fasta file
R (statistical language) could have been an easier choice. Biostring package reads the file as fasta format and removes all the "wraps" while reading the file.
Here are a few steps to follow:
1. Install R (very easy, available for windows also)
2. Open R and install Biostrings Package as instructed int this link https://bioconductor.org/packages/release/bioc/html/Biostrings.html
3. Then run the following commands
getwd() #Shows you the location of R
setwd("Give the path to your fasta file")
library(Biostrings)
f
If you are familiar with Perl programming, this can be very easy to do.
I see that Upendra Kumar has attached a python script that you can use.
You have to open the script by a text editor, change wrap_test.fa by the name of your file and fix the name of the output file by changing wrap_test_out.fa.
Then open a linux terminal and write "python wrap_fix.py" then press enter.
It should work and generate for you a new file containing what you want.
Alternatively, if you open the actual sequence information on perl, select "edit" in the top left corner ------> line operations ---------->join lines. If you had a continuous sequence and wanted to adjust it to single line FASTA, then you would choose "split lines". Be cautious of gene IDs with no sequence information though, alignment programs like BLAST, Clustal, or T -Coffee will not be able to read the information with blank sequences. I would definitely endorse BioPerl because of it's built in FASTA measures.
Simple script with python
Change "exo.txt" to your file name.
Good luck
Here is one line awk code:
awk '/^>/ {printf("\n%s\n",$0);next; } { printf("%s",$0);} END {printf("\n");}' < input.fasta
You can just play around with "sed" and "tr" and you will get the result.
example:
let's say you have a file "wrap.txt" which look something like this:
>accession1
ATGGCCCATG
GGATCCTAGC
>accession2
GATATCCATG
AAACGGCTTA
just use the commands:
cat wrap.txt|sed -e '/^>/s/$/@/' -e 's/^>/#/'|tr '\n' ' '|sed 's/ //g'|sed 's/@/\n/g'|sed 's/#/\n>/g'|sed '/^$/d' > newfasta.fa
You will get the result in a new file named as newfasta.fa
I hope this helps...
Please reply if it doesn't work.
- Badges
- Science topic
- Similar topics
- Computer Science and Engineering
- Bioinformatics
- Bioinformatic Tools