from this error

09:26:35.186 [Thread-1] [ERROR] read(id(B0479ABXX110308:5:1201:11107:16966) coords(chr1:1009343-1009418) cigar(76M) mate(chr1:1009443) flags(99)) exception: java.lang.StringIndexOutOfBoundsException: Range [0, -1) out of bounds for length 5

The output of samtools view .bam | grep "B0479ABXX110308512011110716966"

The reads seem fine so I'd say that your best bet is to start from scratch and realign your bam files or even go for fastqs if you can, and run oncoanalyser from the beginning using those. It seems that the issue you're having is closely related to this one https://gatk.broadinstitute.org/hc/en-us/community/posts/360057841992-MergeBamAlignment-Error-ArrayIndexOutOfBoundsException and it seems that the realigning was the fix for them. Also, if you could post all of this as an issue on GitHub with all the files you've shared with me, that'd be really appreciated!

I agree there is something not quite right with the input alignments, and it looks as if they are missing UMIs. @Sergio Esteban Echeverría do you have information on how the UMIs in the sequence data were handled?

So what I did originally is taking the original TCGA bams, converting them to fastq and realigning them to the HMF genome. I also tried running the pipeline with the original bams and the fastqs and both return the same markdups error (fastqs after performing alignment)

Thanks, I think I understand the problem. For TSO500 data it is expected that the reads are associated with duplex UMIs. Taking your above example, in the FASTQ format they are required to be as follows where

NTAGAAC+CTAGGCT

are the duplex UMIs:

@B0479ABXX110308:5:1201:11107:16966:NTAGAAC+CTAGGCT 1:N:0:NGCTCATT+TAAGATTA
ACCAAAAGGAAAGGAAGATTCTAGCACCGTACGTCAGCATTAAGAAATCAGGAAACGGGCCGGGCGCGGTGGCTCA
+
FFFFFFFFFF:FFF:FFFFFFFFFFFFFFFFFFFFFFFFFFFFFFFF:FFFFFFFFFFF:FFFFFFFFFFFFFFFF

And after mapping the UMIs will then appear in the QNAME field of the SAM record:

B0479ABXX110308:5:1201:11107:16966:NTAGAAC+CTAGGCT  99  chr1    1009343 60  76M =   1009443 176 ACCAAAAGGAAAGGAAGATTCTAGCACCGTACGTCAGCATTAAGAAATCAGGAAACGGGCCGGGCGCGGTGGCTCA    BACAEEEGGDDEGGDEGDACGEAGFCEF=AAE=AGBGGBACAEHEFEAHCGHBEDE=HDGG=GHD>D>E@EEDCC8  NM:i:0  AS:i:152    nn:i:0  tp:a:P  cm:i:9  s1:i:97 s2:i:0  de:f:0  rl:i:61 MQ:i:56 MC:Z:76M    ms:i:2657

Since your alignments are missing the expected UMI data in QNAME, the above error is triggered during processing with MarkDups

I don't think my original data have UMIs, here is a line of the original bams:

C2BBFACXX130823:3:2106:4526:27715       99      chr1    30314   0       76M     =       30406   168     TCCCTCCGGCATTTTTGGCCCTTGCCTTTTAGGGTTGCCAGATTAAAAGACAGGATGCCCAGTTAGTTTGAATTTT      ADDEDED=DECBCDDDDEEDDDDFDEDDDD@HDC>BEEECECBDAEEEGBCCEFCAEFEECFAB?EBBBFC@?BCB    MD:Z:1G60C13    PG:Z:MarkDuplicates     RG:Z:C2BBF.3    NM:i:2  OQ:Z:CCCFFFFFGHHHHJIJGIIIIJJIIIJJJJGJGG?FDHIIGIJIHIJJIEFHGIIGGJHHHHCEBCDDFEE@CEEC AS:i:69 XS:i:69

Within oncoanalyser MarkDups is configured to always expected UMIs in TSO500 data. MarkDups can be run separately with a different configuration (i.e. not to required UMIs)

Just clarify that my data is not TSO500, it is WES from TCGA projects with different target regions capture kits, so I built the resource files for each

Just clarify that my data is not TSO500

Ah thanks! Slipped my mind

Nice! That seems to solve the markdups issue. Thanks for the efficient help!