Make HMDB grouping optional, grouping ppm user-set?

[joannawolthuis]

Hi again :)

Although I have a pipeline in development I'm running into issues(and lack of validation/time to do so) and would love to try yours. I however really would like to skip the HMDB grouping step and just do the 'rest'/untargeted grouping for all the samples - would it be possible to make that toggle-able in the run script?
Additionally, for the grouping step as well, would you consider allowing users to define the grouping ppm?

The scripts look really nice and since we share the HPC I anticipate they will be not too hard to install :D

Cheers,
Joanna

[joannawolthuis]

I got through the bug, using:
https://github.com/compomics/ThermoRawFileParser/
Still interested in the above to do some comparisons, but no hurry :)
That raw file converter worked in one go (miraculously!!) on our HPC using guix and mono <3 So if you ever run into issues, I'd consider it!

[nienkevanunen]

Hi Joanna,

Very nice find! I have a lot of issues with the singularity docker I was using for the conversion, especially when the HPC server is under heavy load by other people already. This method seemed to go very flawlessly indeed (for now). Hopefully not too good to be true. 😛

The ppm shouldn't be too hard to make user-defined (surprised it isn't already).

Skipping steps would be a bit harder, since they all depend on each other. I'll try to look into that if I have the time. Every step's output is kept in the output folder, so maybe just running it as is you will still find some intermediate output that you can use.

[joannawolthuis]

Hi !! I just found out something ( I think ), I think the default output setting is centroided, so the thermo library does the peak calling for you - if you want to do the peak fitting etc. as in the DIMS pipeline, I think it should be turned off with the -p setting! Please check this as it could have big implications I think ._. I was wondering why my peaks looked so weird...