We've moved!
This site is now read-only. You can find our new documentation site and support forum for posting questions here.
Be sure to read our welcome blog!

Problem in using TargetRealigner on bowtie mapped reads

Dear Sir/Ma'am,

I have Exome data from SOLiD 5500xl for which bowtie-0.12.8 was utilized for mapping. The SAM file obtained from bowtie does not have RG tag for every read and also the mapping Quality is 255 for the mapped reads both of which are unacceptable by GATK. By reading on GATK forum I tried appending the RG tag using picard AddorReplaceReadGroups utility which worked fine. Then I tried solving the mapping quality problem by using ReassignMappingQualityFilter but it does not work for me.

The command I used was:
java -jar GenomeAnalysisTK.jar -T RealignerTargetCreator -I bl_mark_duplicate.bam -R /share/reference/human/samtools/hg19.fa -rf ReassignMappingQuality -DMQ 60 -o bl_input.bam

But it does not reasign the mapping quality instead gives me the error of missing quality values.

I solved this problem by replacing the mapping quality value through perl script. After this again I tried using TargetRealigner, though it does not give me error but does not print any thing in output file.

If I just ignore this step and try to do baserecalibration then I get the following errror:

SAM?BAM file is malformed. unable to find color space information in Solid reads. Unfortunately this BAM file can not be recalibrated due to potential reference bias.

Please help me solve this problem. Hoping for a positive response.

Thanks and Regards,
Neha

Answers

  • Geraldine_VdAuweraGeraldine_VdAuwera Cambridge, MAMember, Administrator, Broadie admin

    Hi Neha,

    Can you tell me what version of GATK you are using and post the full error message you get (about missing quality values)?

  • sanghinehasanghineha Member
    edited April 2013

    Hello Geraldine,

    I am using GATK-2.3.9. The complete error message I get is:
    20057754 reads were filtered out during traversal out of 20057754 total (100.00%)
    INFO 15:37:51, 450 MicroScheduler -- > 20057754 reads(100.00% out of total) failing MappingQualityUnavailableFilter.

    Thanks and Regards,

    Neha

  • Geraldine_VdAuweraGeraldine_VdAuwera Cambridge, MAMember, Administrator, Broadie admin

    I see. Can you post a couple of lines from your original BAM file?

  • emmaberdanemmaberdan Member

    Hi I also have this problem. My error message reads: MicroScheduler - 138081698 reads were filtered out during traversal out of 138081698 total (100.00%)
    INFO 21:26:29,995 MicroScheduler - -> 138081698 reads (100.00% of total) failing MappingQualityUnavailableFilter

    The first few lines of my BAM file look like this (I only have two samples)

    QSEQ155.171531443 153 comp0_c0_seq1 103 255 31M * 0 0 CCGAAGGCTCGCGGGTTCGAGTCCCGCTTCG AC9DFA5'?8)))1))C<+2?<<)))=B;88 XA:i:2 MD:Z:1T5A4A18 RG:Z:texFemale_bowtie.coordSorted NM:i:3 XS:A:-
    QSEQ155.149500791 153 comp0_c0_seq1 114 255 21M * 0 0 CGGGTTCGAGTCCCGCCTCGG [email protected]((9.)[email protected] XA:i:2 MD:Z:1A14T4 RG:Z:rubFemale_bowtie.coordSorted NM:i:2 XS:A:-
    QSEQ155.24288163 73 comp45_c0_seq1 59 255 101M * 0 0 GTTTGCTTGGTACAAAACCCTAAAGTGAGAGACTTTTAATTCAACGTTCACCTCAGTGGAAAATAATGGGTGGTACGACATTGCTAGCACAATCGTAAGGC CCCFFFFFHHFHHJJJJJIJJJJJJHHIJIJJGIJJJJJJJJJJJIHIJIIIIJJJHJJHIIIIGJIJIIGIHBBCDCBCDDDDDEDDCCCBCCBDDD?A> XA:i:1 MD:Z:9T84T6 RG:Z:rubFemale_bowtie.coordSorted NM:i:2 XS:A:-
    QSEQ155.66553671 137 comp45_c0_seq1 89 255 96M * 0 0 GACTTTTAATTCAATGTTCACCTCAGTGGAAAATAATGGGTGGTACAACATTGCTAGCACAATCTTAAGGCCTTCAGAAGAAGGCACATGATGGCC CCCFFFFFHHHGHJJJIJJJJJJJJJHHIJIIJJJJJJJJDFHGHIJJIIHIGIIHIIJJJJJJIJIJHIIIIHHHHHHFFFDFDEEEDDCDDDDD XA:i:1 MD:Z:14C31G49 RG:Z:rubFemale_bowtie.coordSorted NM:i:2 XS:A:+
    QSEQ155.133471995 153 comp45_c0_seq1 89 255 101M * 0 0 GACTTTTAATTCAACGTTCACCTCAGTGGAAAATAATGGGTGGTACGACATTGCTAGCACAATCGTAAGGCCTTCAGAAGAAGGCACATGATGGCCAAAAT <AACACCCAC?BDDCCCAA>DFFDEHEEEHHHGHHEJGJIIIHGFDIGGFGGGHEJGIIGJIGIHFIHEDCGDCEIJJIIGHGGGCDDDDHFFD>[email protected]< XA:i:1 MD:Z:64T31G4 RG:Z:rubFemale_bowtie.coordSorted NM:i:2 XS:A:-
    @DD;[email protected]@F[email protected]:<@CH&lt;9A<EFC<31?8DF<DGHDFAD9B8
    ?<FGFHICFGGF=@[email protected]@7?DCCCCC>>:(5; XA:i:1 MD:Z:12C31G49 RG:Z:rubFemale_bowtie.coordSorted NM:i:2 XS:A:-
    @ XA:i:1 MD:Z:62T31G4 RG:Z:rubFemale_bowtie.coordSorted NM:i:2 XS:A:-
    QSEQ155.111635471 153 comp45_c0_seq1 125 255 97M * 0 0 TGGGTGGTACGACATTGCTAGCACAATCTTAAGGCCTTCAGAAGAAGGCACATGATGGCCGAAATGTCAGCCCGTCCGACTGTGATGCGGTTCATTC 7BBADDDBBBDC>ADC5(C>5<[email protected]@DDB?>;@6&gt;[email protected]@;A;;7)7;E=77=:;[email protected];D>HFB86AGGFD6CAFC<[email protected]??HDDFDD??< XA:i:0 MD:Z:97 RG:Z:texFemale_bowtie.coordSorted NM:i:0 XS:A:-
    QSEQ155.124388216 153 comp45_c0_seq1 125 255 97M * 0 0 TGGGTGGTACGACATTGCTAGCACAATCTTAAGGCCTTCAGAAGAAGGCACATGATGGCCGAAATGTCAGCCCGTCCGACTGTGATGCGGTTCATTC DBDCDDDBBDDACCCCCDCCCC>CCA>>[email protected]=DDA>[email protected];?;;[email protected]>[email protected]:[email protected] XA:i:0 MD:Z:97 RG:Z:texFemale_bowtie.coordSorted NM:i:0 XS:A:-
    QSEQ155.52392331 153 comp45_c0_seq1 125 255 100M * 0 0 TGGGTGGTACGACATTGCTAGCACAATCTTAAGGCCTTCAGAAGAAGGCACATGATGGCCGAAATGTCAGCCCGTCCGACTGTGATGCGGTTCATTCTAG DDDDDDDDDDEDDCCDDCDDDDDDDDDDDDDDDDDDDDDDEEEEEEEFFFFFFHGFHJJJJJJJIJJJIJJJJJJJJJJJJJJJJJJHHHHHFFFFFCCC XA:i:0 MD:Z:100 RG:Z:texFemale_bowtie.coordSorted NM:i:0 XS:A:-
    QSEQ155.49827967 99 comp45_c0_seq1 128 255 57M = 230 163 GTGGTACGACATTGCTAGCACAATCGTAAGGCCTTCAGAAGAAGGCACATGATGGCC :==AA7,)[email protected]?4A<C73<+2211)):?70?::0000((9=A<7>BB=7 XA:i:1 MD:Z:25T31 RG:Z:texFemale_bowtie.coordSorted NM:i:1 MQ:i:255 XS:A:-

  • Geraldine_VdAuweraGeraldine_VdAuwera Cambridge, MAMember, Administrator, Broadie admin

    That's because the reads are getting filtered out (due to mapq 255) before the reassignment filter can be applied on the fly. What you should do is use PrintReads with the reassignment filter to generate a copy of your file with reassigned mapqs. Then you can use that new file for processing.

  • emmaberdanemmaberdan Member

    It worked! Thanks!

  • vyellapavyellapa Member

    @neha bowtie-0.12.8 is not a gapped aligner. How does it help to realign indels?

Sign In or Register to comment.