ccms-ucsd / hpp_inspector Goto Github PK

There's a bug when a mod is on the second to last amino acid of the spectrum, the closing bracket is after the final amino acid, not before it. For example: mzspec:PXD010154:01284_D01_P013188_B00_N04_R1:scan:38131:TTGIVLDSGDGVTHNVPIYEGYALPHAIM[+15.994915R]/3

Notes in Slack - to be transferred here

https://github.com/benpullman/peptide_statistics_hpp/blob/501cfff6aaefb93b0de9aa13d06e2aae284cb469/tools/peptide_statistics_hpp/peptide_protein_cosine.py#L450

Use code from PEPTIDE_STATISTICS2 to do the exon mapping at the same time as the database mapping.

https://proteomics2.ucsd.edu/ProteoSAFe/result.jsp?task=2ed077cfc22f4330a394a2de8b8cf3f8&view=view_added_psm#%7B%22protein_input%22%3A%22P00751%22%2C%22table_sort_history%22%3A%22score_dsc%22%2C%22usi_input%22%3A%22mzspec%3AMSV000083957%3AES_XP_LMW_DS_118%3Ascan%3A6047%3AVSEADSSNADWVTK%2F2%22%7D

Is a good match, but has a terrible cosine because of noise.

1. Consider best EI to show, not EI for representative
2. Output all high scoring PSMs, ie best per category

Make sure that this workflow can input a library and output code that Explorer suite outputs (and add resulting result views for double checking the information).

Currently a lot of unused outputs, clean up code to only use a rewritten version of peptide_stats_coverage_map.

Add scores for proteins from inputs and also output protein FDR for specific subsets of proteins (with score cutoff, with 2+ HPP sequences, etc.)

Need to remove SNR2 filter and window filter for annotated cosine, to see.

Read these from ProteinExplorer and use in the output for mapping - seems like super low hanging fruit.