smith-chem-wisc / metamorpheus Goto Github PK

stuff I'd like added or changed:

• Add filename to each scan
• Separate protein accession and protein description
• Include sequence without PTMs
• Mark contaminants
• Add delta score
• Add intensity
• Add retention time
• Include protein location (start residue at least, stop if desired)
• Include units for precursor mass error

Split into ptm, adducts, glyco, unknown, etc. This is for different kinds of runs.

Use mzLib, and the included class generated from schema.

And if it finds too many intervals with multiple peaks, should give a meaningful error return.
Might be wise to limit the tolerance here to plus/minus 1

Even with defects

Need more complicated logic for seen peptides

• Readme
• Wiki

Different ion types z,c, a, etc.

not necessarily "unique" in the sense that they match only one protein, but unique in the sense that the best scoring version of each peptide is kept and all others are dropped. This obviates grade inflation in FDR when multiple copies of the same thing are observed.

use the same output header as psms_tsv

Write the software version in the written files. Decide which of the output files need to have the version.

Have MS1 scan, then ms2 scans are mixed: might be cid, hcd, etc -> all from same precursor

Write the results summaries, decide if need to write results for each engine/task/global task. Probably all of them.

Display tasks in treeview

Look for peptide evidence across multiple spectra files based on m/z and retention time, not requiring MS2 precursor selection. See moff algorithm

Put in a boolean switch in tasks

Dictionary<CompactPeptide, HashSet> local = compactPeptideToProteinPeptideMatching.ToDictionary(b => b.Key, b => new HashSet());

in analysis engine, method AddObservedPeptidesToDictionary
with error

An exception of type 'System.InvalidOperationException' occurred in mscorlib.dll but was not handled in user code

Additional information: Collection was modified; enumeration operation may not execute.