extract_peprophet_data: Get some data from a peptideprophet run.

In elsayed-lab/hpgltools: A pile of (hopefully) useful R functions

Get some data from a peptideprophet run.

Description

I am not sure what if any parameters this should have, but it seeks to extract the useful data from a peptide prophet run. In the situation in which I wish to use it, the input command was: > xinteract -dDECOY_ -OARPpd -Nfdr_library.xml comet_result.pep.xml Eg. It is a peptideprophet result provided by TPP. I want to read the resulting xml table and turn it into a data.table so that I can plot some metrics from it.

Usage

extract_peprophet_data(pepxml, decoy_string = "DECOY_", ...)

Arguments

pepxml	The file resulting from the xinteract invocation.
decoy_string	What prefix do decoys have in the data.
...	Catch extra arguments passed here, currently unused.

Value

data table of all the information I saw fit to extract The columns are: * protein: The name of the matching sequence (DECOYs allowed here) * decoy: TRUE/FALSE, is this one of our decoys? * peptide: The sequence of the matching spectrum. * start_scan: The scan in which this peptide was observed * end scan: Ibid * index This seems to just increment * precursor_neutral_mass: Calculated mass of this fragment assuming no isotope shenanigans (yeah, looking at you C13). * assumed_charge: The expected charge state of this peptide. * retention_time_sec: The time at which this peptide eluted during the run. * peptide_prev_aa: The amino acid before the match. * peptide_next_aa: and the following amino acid. * num_tot_proteins: The number of matches not counting decoys. * num_matched_ions: How many ions for this peptide matched? * tot_num_ions: How many theoretical ions are in this fragment? * matched_ion_ratio: num_matched_ions / tot_num_ions, bigger is better! * cal_neutral_pep_mass: This is redundant with precursor_neutral_mass, but recalculated by peptideProphet, so if there is a discrepency we should yell at someone! * massdiff How far off is the observed mass vs. the calculated? (also redundant with massd later) * num_tol_term: The number of peptide termini which are consistent with the cleavage (hopefully 2), but potentially 1 or even 0 if digestion was bad. (redundant with ntt later) * num_missed_cleavages: How many cleavages must have failed in order for this to be a good match? * num_matched_peptides: Number of alternate possible peptide matches. * xcorr: cross correlation of the experimental and theoretical spectra (this is supposedly only used by sequest, but I seem to have it here...) * deltacn: The normalized difference between the xcorr values for the best hit and next best hit. Thus higher numbers suggest better matches. * deltacnstar: Apparently 'important for things like phospho-searches containing homologous top-scoring peptides when analyzed by peptideprophet...' – the comet release notes. * spscore: The raw value of preliminary score from the sequest algorithm. * sprank: The rank of the match in a preliminary score. 1 is good. * expect: E-value of the given peptide hit. Thus how many identifications one expect to observe by chance, lower is therefore better * prophet_probability: The peptide prophet probability score, higher is better. * fval: 0.6(the dot function + 0.4(the delta dot function) - (the dot bias penalty function) – which is to say... well I dunno, but it is supposed to provide information about how similar this match is to other potential matches, so I presume higher means the match is more ambiguous. * ntt: Redundant with num_tol_term above, but this time from peptide prophet. * nmc: Redundant with num_missed_cleavages, except it coalesces them. * massd: Redundant with massdiff * isomassd: The mass difference, but taking into account stupid C13. * RT: Retention time * RT_score: The score of the retention time! * modified_peptides: A string describing modifications in the found peptide * variable_mods: A comma separated list of the variable modifications observed. * static_mods: A comma separated list of the static modifications observed.

elsayed-lab/hpgltools documentation built on May 9, 2024, 5:02 a.m.