Reorganize outputs

[lovettse]

It would be nice to standardize where outputs land between the various NBFAC tools, so that if I run MAnaT and Yeat on the same sample(s) the directories have a coherent structure and fastp doesn't end up running twice.

My suggestion would be to standardize on analysis/{sample}/{tool} as the location for analysis outputs and seq/{sample}/flt_{sample}_[12].fastq for filtered reads.

[adambazinet]

My concern would be if e.g. fastp is called or parameterized differently workflow to workflow, that you can’t just rely on existence of output in order to automatically skip that step. Maybe there is some way of verifying the commands versions etc exactly? In general though I think the idea of consistent directories structures and so forth is a great idea. Fine for YEAT to be called standalone but I think many envision it called from MAnaT eventually. When that integration is performed it should be possible to remove redundancy.