Workaround in case raw fastqs are gone and new contrasts are needed.

[tovahmarkowitz]

Goal: to make it possible to run new contrasts without using the chrom-seek run command (so as to avoid needing the raw fastqs)

Plan:
Have a script that takes in the current config.json, a new peakcall file, and a new contrast file and produce an updated config.json so that a user can just run the kickoff script without involving the chrom-seek run command

[rroutsong]

Are you basically asking for a run-mode that starts at peak bed files?

So input bed files and get the same outputs from the respective pipelines (chip, atac, cfchip, cutnrun)?

[tovahmarkowitz]

Hi Ryan,

The point here is that there are plenty of times where we want to run new contrasts months/years after the original analysis and to save space we've blown away the raw fastq folder, but left the remainder of the folder intact. The idea would be to basically have it be able to start with the Q5DD bam (or tagAlign) and try to propagate from there as needed.

The wrapper script around kickoff.sh (which also updates config.json) currently checks if the raw fastqs exist. So the idea is to have a script that can update the config.json to allow us to directly use kickoff.sh without requiring the wrapper.

Tovah

[rroutsong]

We can add some logic where-by if you list bams in chrome-seek run --input ..., instead of the traditional .fastq[.gz]?, then we pick up each individual pipeline mode from that point. Does this accomplish what you want?