Tweet
Crossbow alignment format
I am using crossbow v1.1.1 on a local computer cluster to map a set of human SOLiD reads and call the variants. Crossbow glues together the bowtie aligner and SOAPsnp variant caller, and does this in a highly parallel fashion with an option to use the Amazon cloud.
Currently, the program works well and the final SNP output is what is expected. However, I also need the alignment information from the initial bowtie alignment in BAM format, which crossbow does not currently support. Crossbow alignment output files can be retained and are in a 10-column tab-delimited format, but the format does not match standard bowtie output or SOAPsnp input. The following are four lines from the crossbow alignment file:
19 0061 061971493 + TGAGTGAATTACATCATATCCACACAGATGAATGTATTACATCATATC `aXXc__^^_NR^[_`ac_^^ZMLZRTVSOLU?!<S8!+O?!6WS@1- 0 34:A>T 0 r
16 0005 005054723 - GCATGTGTGTGTGTGTGTGTGTGTGTGTGTGTGTGTGGTGTGTGTGAG !!!!GHEFOM>17?ILDDB?BLUA579JRC3:G@JYYJLa_^__TQ]` 0 - 0 r
2 0051 051640947 - TTGGCATCCCAAAGTGCTGGGATTACAGGCATGAGCCACCGTGCCTGG '/5L^;<]PE@@7AFDX][VZ^^ac`[UYbVV``ZZb`_ab_Y\__ba 0 - 0 r
19 0055 055523009 + GTGAGCCGAGATCGCACCACTGCACTCCAGCCTGGGCGACAGAGTGAG ccccccccccbb_XZ]YUYab``ba[]aVTY[]QU_XZIBXacYW`aA 91 - 0 r
Some of these columns are obvious and others are not (at least to me). From what I can gather the fields are:
1: chromosome ID
2: ??
3: start position in reference
4: strand orientation
5: read bases
6: quality string
7: ??
8: variant call (0-based)
9: ??
10: ?? It's always "r" ??
My goal is write a script to turn this file into a SAM file and then use SAMtools for other analyses. Any help on this would be greatly appreciated. Thanks.
Currently, the program works well and the final SNP output is what is expected. However, I also need the alignment information from the initial bowtie alignment in BAM format, which crossbow does not currently support. Crossbow alignment output files can be retained and are in a 10-column tab-delimited format, but the format does not match standard bowtie output or SOAPsnp input. The following are four lines from the crossbow alignment file:
19 0061 061971493 + TGAGTGAATTACATCATATCCACACAGATGAATGTATTACATCATATC `aXXc__^^_NR^[_`ac_^^ZMLZRTVSOLU?!<S8!+O?!6WS@1- 0 34:A>T 0 r
16 0005 005054723 - GCATGTGTGTGTGTGTGTGTGTGTGTGTGTGTGTGTGGTGTGTGTGAG !!!!GHEFOM>17?ILDDB?BLUA579JRC3:G@JYYJLa_^__TQ]` 0 - 0 r
2 0051 051640947 - TTGGCATCCCAAAGTGCTGGGATTACAGGCATGAGCCACCGTGCCTGG '/5L^;<]PE@@7AFDX][VZ^^ac`[UYbVV``ZZb`_ab_Y\__ba 0 - 0 r
19 0055 055523009 + GTGAGCCGAGATCGCACCACTGCACTCCAGCCTGGGCGACAGAGTGAG ccccccccccbb_XZ]YUYab``ba[]aVTY[]QU_XZIBXacYW`aA 91 - 0 r
Some of these columns are obvious and others are not (at least to me). From what I can gather the fields are:
1: chromosome ID
2: ??
3: start position in reference
4: strand orientation
5: read bases
6: quality string
7: ??
8: variant call (0-based)
9: ??
10: ?? It's always "r" ??
My goal is write a script to turn this file into a SAM file and then use SAMtools for other analyses. Any help on this would be greatly appreciated. Thanks.
Comment