findProteinFeatures: Protein feature detection

In CCprofiler/CCprofiler: Complex feature detection using SEC protein chromatograms.

Description

Run the sliding window algorithm to find protein features.

Usage

findProteinFeatures(traces, corr_cutoff = 0.95, window_size = 12,
  parallelized = FALSE, n_cores = 1, collapse_method = "apex_only",
  perturb_cutoff = "5%", rt_height = 5, smoothing_length = 9,
  useRandomDecoyModel = TRUE)

Arguments

traces	An object of class traces (type "peptide").
corr_cutoff	Numeric, the correlation value for chromatograms above which peptides are considered to be coeluting, default=0.95.
window_size	Numeric, size of the window (in fractions), default=12
parallelized	Logical, wether the computation should be done in parallel, default=FALSE
n_cores	Integer, the number of cores to use for parallel processing (only applies if parallelized is TRUE), default=1
collapse_method	Method for collapsing multiple features into one feature: "apex_only": collapses by apex "apex_network": collapses by apex and connected network cluster Default="apex_only"
perturb_cutoff	Numeric, the quantile to use in estimating the perturbation level, default="5 Intensity values that are zero are replaced with random values that are below the specified quantile of the input values. Alternatively a cutoff value can be specified as an upper limit for perturbation values. This is nescessary for correlation calculation.
rt_height	Numeric, RT cutoff for collapsing features, default is 5
smoothing_length	Numeric, smoothing length of Savitzky-Golay filter, default is 7
useRandomDecoyModel	Logical, wether random peptide protein associations should be used as decoy model, default = TRUE

Value

A data.table containing protein features.

Examples

## Load example data
peptideTraces <- examplePeptideTracesFiltered
## Subset traces for shorter processing time
testProteins = unique(peptideTraces$trace_annotation$protein_id)[1:5]
peptideTracesSubset = subset(peptideTraces,trace_subset_ids = testProteins, trace_subset_type = "protein_id")

## Perform co-elution signal detection
proteinFeatures <- findProteinFeatures(traces=peptideTracesSubset)

## Inspect complex features
head(proteinFeatures,n=3)

CCprofiler/CCprofiler documentation built on May 19, 2021, 7:40 p.m.