stephenturner/oneliners

Useful bash one-liners for bioinformatics.

oneliners's Issues

Faster option for "Untangle a FASTQ file. If a FASTQ file has paired-end reads intermingled"

Nice to see this "blog"! Thanks for sharing!

fastq_deinterleave.sh contents:

!/bin/bash

echo "deinterleave_fastq.sh < infile f.fastq r.fastq"
paste - - - - - - - - | tee >(cut -f 1-4 | tr "\t" "\n" > $1) | cut -f 5-8 | tr "\t" "\n" > $2

This is probably the fastest way to do this. ( This is not my own code, got it from somebody else, but I think its nice to share)

Performance differences on 1 1 GB fastq file
seqtk: 30 seconds
alternative:17 seq.

For fastA deinterleave, change the code from 1-4 to 1-2 and from 5-8 to 3-4.

Command for listing everything that does not match a patter requires extended globing be enabled

The one-liner for listing or removing all files that do not match a pattern from the etc section requires that extended globbing be enabled (shopt -s extglob).

It may be a good idea to include this as a note since not all distros have this enabled by default. (mine didn't)

For more info, see here

Awk search command is not working

Print each line where the 5th field is equal to ‘abc123’:

awk '$5 == "abc123"' file.txt
It doesnt work

WHile following command works confirming that data is present in file

Print each line where the 5th field is equal to ‘abc123’:
awk '/abc123/' file.txt

bug in sed 's/[ \t]*$//' example

sed 's/[ \t]*$//' file.txt 

for

cat -e test.txt
bla bla t   $

will remove "t" character

sed "s/[[:space:]]*$//" test.txt | cat -e

works correctly

PDF creation script

Would you consider adding a script with the PDF creation command (probably using PANDOC)?

It may be useful to other people who want to create small informational repos like yours.

add datamash examples

http://www.gnu.org/software/datamash/examples/

H

bam to fastq one liner

Hey,

Thanks very much for putting these explanations and tools up. I think the one liner you have put for converting bam to fastq is inappropriate (or should be described differently). The problem is that your awk prints fields 1, 10, and 11 in the bam.

Field 10 is called SEQ and represents the query sequence to which the read is aligned. However alignment sequences are always represented on the plus strand of the reference (http://chagall.med.cornell.edu/NGScourse/SAM.pdf, http://genome.sph.umich.edu/wiki/SAM), meaning that for stranded bams this tool is inappropriate.

Thanks,

Luke

Add aliases from alias.sh

Lots of good ones here http://alias.sh/

add sed to count the occurrences of each symbol in a fasta file

Hi,
maybe it would be interesting to add how to count the occurrences of each symbol in a fasta file:

sed '/^>/d;s/\(.\)/\1\n/g' file.fa | sort | uniq -ic

Best!