Converting GEO database TXT format to fasta

[Etherella]

Hello!

I'm new to bioinformatics, but I need to perform an analysis of some sort.

I've downloaded data from GEO database it's a large TXT file consisting of many lines that look this way : SCS_0004:2:1:1053:18066#0/1 AGCAATATTGACTACANCCTCATCAAAGCCTGTAGGCACC [YITQR]MST\WN\\TEQU[`]WU]]WPYXXXOXU]`\W` 5 29 29 chr17:68048647-68172163_36129 3979 + 1 1

I need to align those short sequences to a mouse chromosome, and I'm using bowtie under windows.

But the problem is , bowtie doesn't work with this format, can you recommend an easy-to-use tool for windows to convert this format into fasta or just raw?

[vivek_]

This could be done with the unix command line but it would be helpful if you can post a few lines enclosed within code brackets like

paste here

to get a precise idea of the file format.

[Etherella]

This could be done with the unix command line but it would be helpful if you can post a few lines enclosed within code brackets like

paste here

to get a precise idea of the file format.

SCS_0004:2:1:1053:18066#0/1	AGCAATATTGACTACANCCTCATCAAAGCCTGTAGGCACC	[YITQR]MST\WN\\TEQU[`]WU]]WPYXXXOXU]`\W`	5	29	29	chr17:68048647-68172163_36129	3979	+	1	1
SCS_0004:2:1:1053:18066#0/1	AGCAATATTGACTACANCCTCATCAAAGCCTGTAGGCACC	[YITQR]MST\WN\\TEQU[`]WU]]WPYXXXOXU]`\W`	5	29	29	chr17:68048647-68172163_36130	3979	+	1	1
SCS_0004:2:1:1053:18066#0/1	AGCAATATTGACTACANCCTCATCAAAGCCTGTAGGCACC	[YITQR]MST\WN\\TEQU[`]WU]]WPYXXXOXU]`\W`	5	29	29	uc008dkh.1	4033	+	1	1
SCS_0004:2:1:1053:18066#0/1	AGCAATATTGACTACANCCTCATCAAAGCCTGTAGGCACC	[YITQR]MST\WN\\TEQU[`]WU]]WPYXXXOXU]`\W`	5	29	29	chr17:68046720-68172163_36128	3943	+	1	1
SCS_0004:2:1:1053:18066#0/1	AGCAATATTGACTACANCCTCATCAAAGCCTGTAGGCACC	[YITQR]MST\WN\\TEQU[`]WU]]WPYXXXOXU]`\W`	5	29	29	chr17:68046720-68172163_36127	3943	+	1	1
SCS_0004:2:1:1054:5070#0/1	TTTCTCTGTCTTGTCCNCCTAGTTTCCCTCCTGTAGGCAC	aaaaaaaaaaaaaaa]EaaaW]]]Yaa\a`[aa]Pa^]VT	2	30	30	chr2:40378133-40378584_42654	395	-	1	1
SCS_0004:2:1:1054:5070#0/1	TTTCTCTGTCTTGTCCNCCTAGTTTCCCTCCTGTAGGCAC	aaaaaaaaaaaaaaa]EaaaW]]]Yaa\a`[aa]Pa^]VT	2	30	30	chr1:8926487-8927380_99	16	+	1	1

Something like this. By the way, I don't have unix installed only Mac OS and windows, but as far as I understand Mac OS is a unix-based system, right?

[dpryan]

The easiest way would be to write a small script (in python, perl, whatever) to read that in and spit out the same data (sans alignment information) in fastq format. Column 1 is the read id, column 2 is the sequence, and column 3 is the quality score. If you have python installed on your Mac, then the following would probably work (changing INPUT_FILENAME to the name of the file you got from GEO and SOME_OUTPUT_FILE to whatever you want the output to be):

#!/usr/bin/python
import csv

f = csv.reader(open("INPUT_FILENAME", "r"), dialect="excel-tab")
output = open("SOME_OUTPUT_FILE", "w")

last = ""
for line in f :
    if(line[0] != last) :
        output.write(">%s\n" % (line[0]))
        output.write("%s\n" % (line[1]))
        output.write("+\n") 
        output.write("%s\n" % (line[2]))
        last = line[0]
output.close()

Something like that would probably work.

[Etherella]

thanks for the reply, I managed to get it working through galaxy.