README.md

In frhl/pRoteomics: Proteomics analysis using R

pRoteomics

Description

Downstream analysis of proteomics data for iTRAQ (labelled data). Modular workflow, i.e. preprocess and analysis. Some of the functions will attempt to guess column nanmes if nothing have been specified. These tools have only been tested on data generated from the Whitehed Institution, Cambridge MA.

workflow

The following steps can be accomplished in pipeline:

Preprocess

using prepare.R: - log2 transformation - median normalization - filter out proteins from unwanted species - filter out protein which fragments have only been observed once. - calculate log fold change

Analysis

using genoppi.R - create replicate plots - create volcano plots - look for protein-protein interactions - user hypergeometric (fisher exact test) to test for enrichment

note

Tools have not yet been fully tested.

example on how to setup personal pipeline

library(rProteomics)

infile = 'data/raw/frederik/frederik_BCAS3_proteins.csv' prelim <- prepare(c("EC", "BCAS3"), infile = infile, impute = list(shift = -0.8, stdwidth = 0.3)) known.interactors <- interactors("BCAS3") # get known interactors

data <- prelim %>% mttest() data %>% designate(FDR < 0.1) %>% plotScatter(bait, paste(bait, '[FDR < 0.1]')) data %>% designate(FDR < 0.1) %>% plotVolcano(bait) data %>% designate(FDR < 0.1) %>% plotOverlap(bait, known.interactors)

frhl/pRoteomics documentation built on Feb. 19, 2020, 3:53 p.m.