If providing a samplesheet check it matches the files (.bam or .fastq) the pipeline recognises at launch.

[fgponce]

Is your feature related to a problem?

My bad having all my demultiplexed .bams in one folder.
I have spent a lot of time troubleshooting resource allocation regarding RAM and timing out of jobs on the cluster. Presumably fastcat was making one huge 300 GB file and processing that in one go. I learnt a lot about setting configs etc but turned out to be that I had all my demultiplexed .bams in the same folder. I know you mention this in the -h input commentary but a quick check at launch would have saved me a lot of time (and I just found another issue thread that did the same).

Describe the solution you'd like

See title.

Describe alternatives you've considered

Also, only one sample being processed (seen by the pipeline) but with the de_analysis flag in the command could throw an error early on and catch this?

Additional context

No response

[cjw85]

Deleted malicious comments.

@fgponce This is a bit of a chicken and egg scenario. One reason the sample sheet is requested is to help the code at the front-end of the workflow make sense of the data it is presented. If you can provide logs and an detailed commentary of the layout of the data you provided then we can begin to think what more might be done to shortcut futile work.