convert Eland sorted.txt to bam file

I downloaded a human ChIP-seq data set mapped to hg18  from NCBI GEO.

it is in Eland sorted.txt format, I want to convert it to bam file for some downstream analysis.

a google search: http://www.biostars.org/p/3121/
I followed the link to http://genomewiki.ucsc.edu/index.php/ABRF2010_Tutorial

the export2sam.pl script in the samtool is what I need.

I downloaded samtools-0.1.8 and found the script in the misc folder.

but I got  errors when I tried to convert it:


tommy@tommy-ThinkPad-T420:~/Downloads/samtools-0.1.8/misc$ ./export2sam.pl --read1=test.txt 
@PG ID:export2sam.pl VN:2.0.0 CL:./export2sam.pl --read1=test.txt
Use of uninitialized value $t[21] in string eq at ./export2sam.pl line 279, <$fh1> line 1.
Use of uninitialized value $t[21] in string eq at ./export2sam.pl line 279, <$fh1> line 2.
Use of uninitialized value $t[21] in string eq at ./export2sam.pl line 279, <$fh1> line 3.
Use of uninitialized value $t[21] in string eq at ./export2sam.pl line 279, <$fh1> line 4.
Use of uninitialized value $t[21] in string eq at ./export2sam.pl line 279, <$fh1> line 5.
Use of uninitialized value $t[21] in string eq at ./export2sam.pl line 279, <$fh1> line 6.
Use of uninitialized value $t[21] in string eq at ./export2sam.pl line 279, <$fh1> line 7.
Use of uninitialized value $t[21] in string eq at ./export2sam.pl line 279, <$fh1> line 8.
Use of uninitialized value $t[21] in string eq at ./export2sam.pl line 279, <$fh1> line 9.
Use of uninitialized value $t[21] in string eq at ./export2sam.pl line 279, <$fh1> line 10.

closely looking into the eland export file, I found that it missed a column at the [21] position.

A sample sorted.txt file looks like this:


SOLEXA2 90529 1 1 0 1514 0 1 AGCAGATGTTAAAGCAAGACNAACAGTCTCCC LWYWSWYYWZZZXVZZZXZMDNZZYQYZZZZZ chr21.fa 46467450 R 20C11 84 N
SOLEXA2 90529 1 1 0 32 0 1 CCTAAGCAAAGAAGTCACTCNGTCACTGAACA NHMFPXXUUVYLPSUUUUYMDKVVTWYYUXXX chr21.fa 45472787 F 1T18A11 69 N
SOLEXA2 90529 1 1 5 121 0 1 TCGAGTGCACACACCCAGCAGCTCTGCAGATA \SH_\ZYU_U_\\_]]_]_^a\_X^\^XSS_\ chr21.fa 43555500 F 1A30 69 Y



the last column contains a letter N or Y denoting whether the sequence passed the quality filtering.

So, I have to add a  "Y" to all the lines in my file:


tommy@tommy-ThinkPad-T420:~/Downloads/samtools-0.1.8/misc$ cat test.txt | awk '{print $0, "\tY"}'  > eland.txt

then, use the export2sam script:

tommy@tommy-ThinkPad-T420:~/Downloads/samtools-0.1.8/misc$ ./export2sam.pl --read1=eland.txt    | perl -wpe 's/(chr.*)\.fa/$1/'  > test.sam

#  perl changes all the chr1.fa to chr1  etc....

And this sam file lacks header http://blog.daum.net/seq/20   the headers are tab delimited.

@HD VN:1.0 SO:unsorted
@SQ SN:chr1 LN:247249719
@SQ SN:chr2 LN:242951149
@SQ SN:chr3 LN:199501827
@SQ SN:chr4 LN:191273063
@SQ SN:chr5 LN:180857866
@SQ SN:chr6 LN:170899992
@SQ SN:chr7 LN:158821424
@SQ SN:chr8 LN:146274826
@SQ SN:chr9 LN:140273252
@SQ SN:chr10 LN:135374737
@SQ SN:chr11 LN:134452384
@SQ SN:chr12 LN:132349534
@SQ SN:chr13 LN:114142980
@SQ SN:chr14 LN:106368585
@SQ SN:chr15 LN:100338915
@SQ SN:chr16 LN:88827254
@SQ SN:chr17 LN:78774742
@SQ SN:chr18 LN:76117153
@SQ SN:chr19 LN:63811651
@SQ SN:chr20 LN:62435964
@SQ SN:chr21 LN:46944323
@SQ SN:chr22 LN:49691432
@SQ SN:chrX LN:154913754
@SQ SN:chrY LN:57772954
@SQ SN:chrM LN:16571

So, I "cat" the header to my resulted sam file and convert it to bam with samtools view.