jpayne@1
|
1 #!/bin/bash
|
jpayne@1
|
2 # Usage: deinterleave_fastq.sh < interleaved.fastq f.fastq r.fastq [compress]
|
jpayne@1
|
3 #
|
jpayne@1
|
4 # Deinterleaves a FASTQ file of paired reads into two FASTQ
|
jpayne@1
|
5 # files specified on the command line. Optionally GZip compresses the output
|
jpayne@1
|
6 # FASTQ files using pigz if the 3rd command line argument is the word "compress"
|
jpayne@1
|
7 #
|
jpayne@1
|
8 # Can deinterleave 100 million paired reads (200 million total
|
jpayne@1
|
9 # reads; a 43Gbyte file), in memory (/dev/shm), in 4m15s (255s)
|
jpayne@1
|
10 #
|
jpayne@1
|
11 # Latest code: https://gist.github.com/3521724
|
jpayne@1
|
12 # Also see my interleaving script: https://gist.github.com/4544979
|
jpayne@1
|
13 #
|
jpayne@1
|
14 # Inspired by Torsten Seemann's blog post:
|
jpayne@1
|
15 # http://thegenomefactory.blogspot.com.au/2012/05/cool-use-of-unix-paste-with-ngs.html
|
jpayne@1
|
16
|
jpayne@1
|
17 # Set up some defaults
|
jpayne@1
|
18 GZIP_OUTPUT=0
|
jpayne@1
|
19 PIGZ_COMPRESSION_THREADS=10
|
jpayne@1
|
20
|
jpayne@1
|
21 # If the third argument is the word "compress" then we'll compress the output using pigz
|
jpayne@1
|
22 if [[ $3 == "compress" ]]; then
|
jpayne@1
|
23 GZIP_OUTPUT=1
|
jpayne@1
|
24 fi
|
jpayne@1
|
25
|
jpayne@1
|
26 if [[ ${GZIP_OUTPUT} == 0 ]]; then
|
jpayne@1
|
27 paste - - - - - - - - | tee >(cut -f 1-4 | tr "\t" "\n" > $1) | cut -f 5-8 | tr "\t" "\n" > $2
|
jpayne@1
|
28 else
|
jpayne@1
|
29 paste - - - - - - - - | tee >(cut -f 1-4 | tr "\t" "\n" | pigz --best --processes ${PIGZ_COMPRESSION_THREADS} > $1) | cut -f 5-8 | tr "\t" "\n" | pigz --best --processes ${PIGZ_COMPRESSION_THREADS} > $2
|
jpayne@1
|
30 fi
|