converting consensus fastq to fasta

[maubp]

Are the sequences and quality lines in the original FASTQ line wrapped? This is valid but not recommended as some tools (perhaps including MAQ) don't understand it.

The tools I suggested can all cope with line wrapped FASTQ. See also: http://dx.doi.org/10.1093/nar/gkp1137

(Or it may be the forum making it look line wrapped and adding spaces when it isn't really - trying using the [ code ] text [ /code ] tags to stop this)

[zlu]

Perhaps I didn't explain it clearly. I wasn't trying to convert the Illumina short reads but mapped consensus sequence of a few chromosomes. Each is a few Mbp long, e.g:

@chr1
nnnnnnnagatagaaataCACGATGCGAGCAATCAAATTTCATAACATCACCATGAGTTT
GGTCCGAAGCATGAGTGTTTACAATGTTTGAAtaCCTTATACAGTTCTTATACATACTTT
ATAAATTATTTCCCaagctgttttgatacactcactaacagaTATTCTATAGAAGGAAAA
GTTATCCACTTATGCACATTTATAGTTTTCAGAATTGTGGATAATTAGAAATTACACACA
AAGTTATACTATTTTTAGCAACATATTCACAGGTATTTGACATATAGAGAACTGAAAAAG
TATAATTGTGTGGATAAGTCGTCCAACTCATGATTTTATAAGGATTTATTTATTGATATT
TACATAAAAATACTGTGCATAACTAATAAGCAGGATAAAGTTATCCACCGATTGTTATTA
+
!!!!!!!????????BBBEEEHKKKKKKKKKKKKNNNNNNQQNNNNNNQWWWW7WZWWWZ
ZZZ]]]`````````>]]]]]]]ZTQQQTQNNKKKKKKHKKHHHEEEEEEEHHHHHHHHH
HKKHHHHHHEEEEEBBBBBBBBBBBB?=B@BBBBBB??BBBBEEEEEEEEEEHKNNNNNQ
QQTTTTWWWWWWWTTWZWWW]`>```c``]`ccc``cfcfliiloouSxxuuuuTollLo
olliifilloolfif````]]WWWWWTTTTTTTTQQQQQQQNKKHHHHEEEEHEEEHHEE
EEEEEEEHHHHHHHHHEEEEEEHHHHHEEHKHHHHHHHHHHHEEHHHHHHHHKNNNNKNN
NQQ@NKNNNQQQTWWWZZZBZZZZ]<`]ZZZZZZZZLZZ``]]]ZZZWTTTQQQNNNNNK

And using the method described above, this is what I got:

>chr1nnnnnnnagatagaaataCACGATGCGAGCAATCAAATTTCATAACATCACCATGAGTTT
>`]]``fGff`````]iifc`````cciiLoxuuuxZ{{xxruuxx{{{~~rrrrrrrrr
Ooiiff`ZZZWTTTTTTQQNTQTWWWWWW]]]`]]cfifffciiiiiorrxx{{~x~{{x
>QNNQQQTTQQQQNNNNQZ``cfffolllloollruuuurrroruxxxxx{~~xuroorr
rrorruxxxroooux{{{xxuuuuuruurrrrollccfcc]ZZB]?]]W7QQTTWWWWWW
>QTT7TQQTTTWZZZZ]]]]]]ZZZZ]]]]]]`````]]]]]`]`c`]```Z]``fflll
louxxxxuS{{{{{~~~~~~{{~{{{{xuuroiilroiiollorlloorZ{~~{xxxuuu
>KKKKKKKHHEB????BBBEEKHHHKKKKNNNNQQQQTZZZZZZ]]````````]]WWW<
WWWW7TTTTTTTQKKKKKKHHEEEE<BBEHHH=HEEEEEHHHHEEEEHHHHHHKKKKKKK
>QQQBQQQNNNNKKKKKKQQQQQQTTTTWWZWTTQQQQQQQQTTWZZZZZ]ZZTWWTTTQ
QQTQQNNNTWWWZZ]]]ZZWWWWWWTTTTQQQEQQQQNNNNNNHHHHEEBEEEEEBBBBB
>QQQATTTTTTTWWTTTQQQQQQQNNKKKQQQTTTT7TTQQ5QNNNQKNNQQQQQQQQQQ
QQN4HKKHHHHHHHHHHHHHKHHEEBBBBBBBBBBBBBBB7??????????????!!!!!
>LAcffffooor{{Z~~~~~~~~~~|~~~q~~~~~v~~~~~~~~~~~~~~~~~~~~~~~~
~~~~~~~~~~~~~~c~~xuuric```]]ZZWTNKHEEB????!!!!!!!!!!!!!!!!!!
>QQQ5>QQ@QNNNNNNQQQQNNNNNNKHEBBBB???BBBBBBBBBEEEEEEBBBBBBBBB
BBBBBBEHKNNKKKKKKNNNKKKKKKKKKKKW]]]cccfffc`]Z`>cfiOollllllLl
>NQQTTTWQQQQQQTTNNNQQQT7GTTAWWWTTTZZW7TQTTQQQQQQTQQTTTQTTTTT
WWZWZ]]]]WZ]``D]]]D```D]]]Z]]]ZZWWWT7@QNKKKKKHEBBBBHKHHHNQTT

[maubp]

I was trying to convert fastq consensus sequneces (generated by the pileup2fq utility of samtools) to a multi-fasta file with the MAQ's fq_all2std.pl fq2fa script but it didn't work. The result is a mix of 60bp sequence and quality lines with the > headers.

Is there quick way to extract only the sequences from the fastq file?

Thank you.

I personally use Biopython, e.g.

http://www.biopython.org/wiki/Reading_from_unix_pipes

http://news.open-bio.org/news/2009/09/biopython-convert-function/

from Bio import SeqIO
count = SeqIO.convert("example.fastq", "fastq", "example.fasta", "fasta")
print "Converted %i reads" % count

You might also consider BioPerl or EMBOSS seqret, etc.

However, the problem could be a broken FASTQ file. Could you show use the start of the file (say two reads)?

[zlu]

I was trying to convert fastq consensus sequneces (generated by the pileup2fq utility of samtools) to a multi-fasta file with the MAQ's fq_all2std.pl fq2fa script but it didn't work. The result is a mix of 60bp sequence and quality lines with the > headers.

Is there quick way to extract only the sequences from the fastq file?

Thank you.