I have tried uploading several .bam datafiles, but for some reason they cannot be detected as the bam datatype (and therefore not used in my workflow). I’ve tried manually setting the datatype to bam both before and after uploading the files. But it gives me the error ‘An error occurred setting the metadata for this dataset. Set it manually or retry auto-detection’. Both of these options gave a similar error. The datasets look like the image. I have uploaded .bam files before and it was never an issue. The only difference is that these bamfiles were generated from .cram files. I’m quite novel to working with these datatypes, so hope I’ve given all the information needed to help me with this issue. How can I make sure they are detected as proper bamfiles?
Try this: check the file content before Upload to Galaxy. Try converting bam to sam format with Samtools. You could also try sorting the bam. If it fails for you locally, then it would fail in Galaxy.
Or, try this: directly Upload the cram data to Galaxy. Most tools do not directly use cram format, but you can convert cram to bam in Galaxy with the tool Samtools view.
For either, try letting the Upload tool auto detect the format first. If that doesn’t work, you can assign the datatype instead (during Upload – not sure if you tried that already for the bam or not).
I’m also not exactly sure how good we are at auto detecting cram format right now… but I’ll test that later today or tomorrow since am now curious. If any usage deviates from the above – I’ll post back.
