In hyginn/BCB420.2019.ESA: Exploratory Systems Analysis Tools

nocite: @steipe-rptPlus

knitr::opts_chunk$set(
  collapse = TRUE,
  comment = "#>"
)

# load BCB420-2019.ESA itself for knitr:
pkgName <- trimws(gsub("^Package:", "", readLines("../DESCRIPTION")[1]))
library(pkgName, character.only = TRUE)

 

If any of this information is ambiguous, inaccurate, outdated, or incomplete, please check the [most recent version](https://github.com/hyginn/BCB420.2019.ESA) of the package on GitHub and if the problem has not already been addressed, please [file an issue](https://github.com/hyginn/BCB420.2019.ESA/issues)!

 

About this vignette:

This vignette describes the workflow that was used to prepare the BioGRID dataset for the package. Source data is protein-protein interaction data from BioGRID [@pmid30476227].

BioGRID Data

The BioGRID is a collection of curated protein-protein interaction data. BioGRID data is licensed under the MIT license. This document describes work with BioGRID 3.5.170 (2019-02-25) [@pmid30476227].

 

Data semantics

BioGRID interaction data is available in various formats common to protein-protein interaction databases. For our purposes of working with HGNC symbols, the BioGRID TAB 2.0 file format appears useful. Details are described here.

The file BIOGRID-ALL-3.5.170.tab2.zip contains the following columns of interest to us:

1. Official symbol for Interactor A. A common gene name/official symbol for interactor A. Will be a “-” if no name is available.
2. Official symbol for Interactor B. Same structure as column 8.
3. Experimental System Type. This will be either "physical" or "genetic" as a classification of the Experimental System Name.
4. Organism ID for Interactor A. This is the NCBI Taxonomy ID for Interactor A.
5. Organism ID for Interactor B. Same structure as 16.

 

Data download and processing

1. Navigate to the download directory of the BioGRID database.
2. Download the following data file:
3. BIOGRID-ALL-3.5.170.tab2.zip (70.7 Mb);
4. Uncompress the file and place it into a sister directory of your working directory which is called data. (It should be reachable with file.path("..", "data", "BioGRID")). Warning: ../data/BioGRID/BIOGRID-ALL-3.5.170.tab2.txt is 618.9 Mb!

 

Preparations: packages, functions, files

To begin processing, we need to make sure the required packages are installed:

readr provides functions to read data which are particularly suitable for large datasets. They are much faster than the built-in read.csv() etc. But caution: these functions return "tibbles", not data frames. (Know the difference.)

if (! requireNamespace("readr")) {
  install.packages("readr")
}

 

Selecting protein-protein interactions

FN <- file.path("..", "data", "BioGRID", "BIOGRID-ALL-3.5.170.tab2.txt")
BioGRID <- as.data.frame(readr::read_tsv(FN,
                                         skip = 1,
                                         col_names = c("A",
                                                       "B",
                                                       "type",
                                                       "taxID_A",
                                                       "taxID_B"),
                                         col_types = "-------cc---c--ii-------"))

nrow(BioGRID)  # 1647089

# Remove all interactions for which taxID_A and taxID_B are not both 9606
BioGRID <- BioGRID[BioGRID$taxID_A == 9606 & BioGRID$taxID_B == 9606, 
                   c("A", "B", "type")]
nrow(BioGRID)  # 442753

# Remove all interactions for which A and B are not both in HGNC sym
myURL <- paste0("https://github.com/hyginn/",
                "BCB420-2019-resources/blob/master/HGNC.RData?raw=true")
load(url(myURL))  # loads HGNC data frame
BioGRID <- BioGRID[BioGRID$A %in% HGNC$sym & BioGRID$A %in% HGNC$sym, 
                   c("A", "B", "type")]
nrow(BioGRID)  # 433505

# How many genetic interactions?
sum(BioGRID$type == "genetic")  # 4702

# Save dataset:
saveRDS(BioGRID, file = file.path("..", "data", "BioGRID", "BioGRID.3.5.170.rds"))  # 2.8 Mb

# The dataset was uploaded to the assets server and is available with:
BioGRID <- fetchData("BioGRID")

 

Acknowledgements

 

Session Info

This release of the BCB420.2019.ESA package was produced in the following context of supporting packages:

sessionInfo()

References

hyginn/BCB420.2019.ESA documentation built on May 29, 2019, 1:23 p.m.