I'm working with TCGA WGS BAMs through the Google Cloud Platform, and I've seen that there are files where the barcode is identical, but there is an additional number appended to the filename before the .bam extension. For example,
G32450.TCGA-FD-A3N5-01A-11D-A21A-08.1.bam
G32450.TCGA-FD-A3N5-01A-11D-A21A-08.3.bam
G32450.TCGA-FD-A3N5-10A-01D-A21A-08.2.bam
G32450.TCGA-FD-A3N5-10A-01D-A21A-08.4.bam
Do these correspond to multiple parts of one BAM that should be merged before analysis? Or is this something else?
These 2 are tumour, based on the part that I have put in bold:
G32450.TCGA-FD-A3N5- 01 A-11D-A21A-08.1.bam
G32450.TCGA-FD-A3N5- 01 A-11D-A21A-08.3.bam
These 2 are from normal tissue:
G32450.TCGA-FD-A3N5- 10 A-01D-A21A-08.2.bam
G32450.TCGA-FD-A3N5- 10 A-01D-A21A-08.4.bam
All of these, however, are from the Broad Institute, as judged by the final number 08,
Unless otherwise stated, the assumption, based on the 'rules' of naming samples, is that these are from the same aliquot (1 tumour aliquot; 1 normal aliquot).
Check the BAM [SAM] headers to see if they have also been processed in the same way. The commands used for alignment should be encoded in the header. In some situations, I've spotted TCGA BAMs from the same project aligned to different genomes.
