gatk countcovariate ERROR MESSAGE Read starting with deletion. Cigar: 1D22M1I71M8S

[aituka]

Hello,

I received targeted sequence reads from a 454 machine.
In my way on the application of the GATKV3.0 best practice guideline, I got the following error:

##### ERROR MESSAGE: SAM/BAM file SAMFileReader{/tmp/ISDBM229314/ISDBM229314.marked.realigned.fixed.bam} is malformed: Read starting with deletion. Cigar: 1D22M1I71M8S

I am using bwa0.6.1 hg19 and GATK1.6.2

Here is the pipeline:
1/ alignment using bwa bwasw -->sam
2/ use PICARD AddOrReplaceReadGroups to get an indexed BAM file:

java -Xmx4g -Djava.io.tmpdir=/tmp -jar picard-tools-1.62/AddOrReplaceReadGroups.jar RGID=IDa RGLB=LB RGPL=454 RGPU=pu RGSM=sample SO=coordinate INPUT=sample.sam OUTPUT=sample.bam VALIDATION_STRINGENCY=LENIENT CREATE_INDEX=true

3/ GATK MarkDuplicates
4/ GATK RealignerTargetCreator
5/ GATK IndelRealigner
6/ PICARD FixMateInformation
7/ GATk CountCovariates --> ERROR MESSAGE

java -Xmx4g -jar GATK.jar -l INFO -R h37.fasta -nt 16 -knownSites:dbsnp,VCF dbsnp_135.b37.vcf -I sample.marked.realigned.fixed.bam -T CountCovariates --read_filter BadCigar -cov ReadGroupCovariate -cov QualityScoreCovariate -cov CycleCovariate -cov DinucCovariate -recalFile sample.recal_data.csv

I have the error with or without the option --read_filter BadCigar

any ideas? hints? links?
thank you for your help.
tuka.

[newbie16]

Did you find solution

Hi aituka,

I have the sme problem. Were you able to find a solution? If yes, could you share?

Thanks

[TiborNagy]

Our solution is simple: remove the read.