Julianus Pfeuffer
Julianus Pfeuffer
PeptideIndexer already can do some sophisticated guess, we should parse its output or store the decoy string in the output files to use it e.g. in PercolatorAdapter which cannot guess...
Possible enhancements in the workflow: - make it optional - reduce number of files needed from previous steps (needs issue on the pmultiqc repo) - attach to EMail if possible...
First try to include XTandem as another search engine.
E.g. different enzymes. Note that e.g. Comet can directly search with a second enzyme. The difficulty is that you usually need to specify if the enzyme was applied after each...
Useful for huge studies where the samples are actually unrelated. Unrelatedness could be one or more of the following: - Different number of fractions - Different instruments - Unrelated conditions...
This would introduce a for-loop around ProteomicsLFQ with the contrasts you actually want to test. That: a) you reduce memory consumption significantly since not all maps are loaded and aligned...
We could also overwrite the experimental design generated from the SDRF but I don't know if that gets even more confusing. Maybe in the long run we even deprecated the...
- Address remaining TODOs. - Adhere to nextflow coding standard by prefixing all channels with ch (required by some code highlighters).
E.g. from UniProt, if no fasta is given. Organism can be another parameter or read from the SDRF.