In hancockinformatics/EssentialTnSeq: TnSeq Essentiality Analysis

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

Installation and Setup

The devtools package is required for installation of EssentialTnSeq. Additionally, the functions included with this package are designed to work with (and internally make use of) the tidyverse suite of packages & functions.

# First, install dependencies devtools and tidyverse
# install.packages(c("devtools", "tidyverse"))

# Then install and load EssentialTnSeq, along with tidyverse
# devtools::install_github("travis-m-blimkie/EssentialTnSeq")

# Load required packages with library()
library(tidyverse)
library(EssentialTnSeq)

Introduction

This package was designed to facilitate the analysis of TnSeq data; specifically, the results of Tradis or Transit/Gumbel essentiality pipelines. The package includes three functions which are designed to be used together to create a complete workflow, from Tradis or Transit/Gumbel outputs to lists of essential genes.
The three functions and examples of their usage are demonstrated in this vignette.

Tradis

0. Input Data

As mentioned previously, this package and its functions are built to make use of the output from essenitality analysis of TnSeq data from Tradis or Gumbel.

Input files for the Tradis side of this essentiality pipeline should have been generated as a result of calling tradis_gene_insert_sites from the Tradis suite of functions. Files should be named something like "condition1_genomeRef.tradis_gene_insert_sites.csv.all.csv". Note final element of the file name ("csv.all.csv"), which is used by the et_readFiles() function to identify relevant files, and as such should be consistent with what is denoted here.

An example directory containing Tradis results is included below.

1. Reading Data

This first step is covered by the function et_readFiles(). The required arguments for this function are detailed below, along with an example (keeping with the directory shown above).
Arguments:

tool - one of "Tradis" or "Gumbel". For this current workflow, we will use "Tradis".
conditions - a list or character vector of condition names. These should match file names, and not overlap with one another. For example, if you have bacteria grown in different media, along with treatment/no treatment, they should be named like: "media1_untreated" and "media1_treated". In this case, using just "media1" in place of "media1_untreated" means the incorrect number of files would be identified (as "media1" also matches "media1_treated".)
reps - the number of replicates for the experimental conditions. Should be a character vector, like c("r1", "r2", "r3"). It should also match the number of files for each condition, otherwise file input will not function properly (you will get an error).
data_folder - the directory containing all of the data files to be used (i.e. all conditions and replicates).

The code below utilizes the preview data shown above, reading in the files with the et_readFiles() function and previously described arguments.

my_conditions <- c("", "", "")
my_reps <- c("r1", "r2", "r3")

raw_data <- et_readFiles(tool = "Tradis", 
                         conditions = my_conditions, 
                         reps = my_reps, 
                         data_folder = "")