In CityUHK-CompBio/PAGnet: PAGnet: Pseudomonas aeruginosa genomic integrated regulatory network.

knitr::opts_chunk$set(echo = TRUE, eval = TRUE,
                      fig.align   = 'center',
                      fig.width   = 8,
                      fig.height = 4)

Overview

Regulatory networks including virulence-related transcriptional factors (TFs) determine bacterial pathogenicity in response to different environmental cues. Pseudomonas aeruginosa, a Gram-negative opportunistic pathogen of humans, recruits numerous TFs in quorum sensing (QS) system, type III secretion system (T3SS) and Type VI secretion system (T6SS) to mediate the pathogenicity. Although many virulence-related TFs have been illustrated individually, very little is known about their crosstalks and regulatory network. Here, based on chromatin immunoprecipitation coupled with high-throughput sequencing (ChIP-seq) and transcriptome profiling (RNA-seq), we primarily focused on understanding the crosstalks of 20 virulence-related TFs, which led to construction of a virulence regulatory network named PAGnet (Pseudomonas aeruginosa genomic integrated regulatory network) including 82 crosstalk targets.

The PAGnet uncovered the intricate mechanism of virulence regulation and revealed master regulators in QS, T3SS and T6SS pathways. The package PAGnet is designed for network visualization, subnetwork filtering and master regulator analysis locally by running a local shniy GUI within R. In addition, the master regulator analysis can be performed in the R console without runnning the shiny GUI to provide more flexibility.

Before starting to use this package, you need to install the following packages:

library(shiny)
library(shinythemes)
library(scales)

Then load the package:

  ## loading package

  library(PAGnet)

Quick Start

Master Regulator Analysis

The user can choose to use the default PAGnet or to upload their own regulatory network in a predefined format (Locus Tag).

A dataframe format of default PAGnet with two colmn transcription factors and targets is provided for MRA, user could also their own regulatory network in same format as input.

  data(PAGnet)

  ##Use defaultPAGnet as regulatory network

  head(pagnet)

The users also need to provide a character vector of TFs (or only interested TFs).

  ## define transcription factors in regulatory network

  head(tf)

A character vector of locus tag should be provide as signature genes.

  ##Use QS related genes as signature genes

  head(qs)

The function pagnet.mra is used to perform MRA over provided regulatory network. The MRA computes the overlap between the transcriptional regulatory unities (regulons) and the input signature genes using the hypergeometric distribution (with multiple hypothesis testing corrections). Having completed master regulator analysis, a table will be returned.

  ## Perform MRA and return results

  mra_results <- pagnet.mra(rnet=pagnet,tflist=tf,signature = qs, 
                            pValueCutoff = 0.05)

  mra_results

Local shiny interface

The function pagnet.mra.interface is used to call local shiny GUI (Figure 1).

  ## call local shony interfact
  pagnet.mra.interface()

A screenshot of local shiny interface of PAGnet

The shiny GUI provides visualization and exploration of PAGnet (Figure 2). It allows the user to filter the full network by selecting one or multiple transcription factor(s) to obtain a subnetwork of interest. A brief summary of the subnetwork is also provided with information about the transcription factors and their target genes.

Visualization of PAGnet in local shiny GUI

The local shiny GUI also provides master regulator analysis for identification of key transcription factors mediating a biological process or pathway (Figure 3). First, the user can choose to use the default PAGnet or to upload their own regulatory network in a predefined format. Second, the user needs to specify a gene signature associated with a biological function or pathway of interest, either by selecting a gene set from public databases or uploading a user-customized gene list. In the current version, the platform provides gene sets in Gene Ontology (GO) and KEGG databases obtained from Pseudomonas Genome DB (@winsor2015enhanced). Having completed master regulator analysis, a table will be returned with information about each transcription factor’s corresponding gene ID, gene name, number of target genes, total number of hits (all signature genes in the network), observed hits (signature genes in the TF’s regulon), and a p-value calculated based on a hypergeometric test. The table is sorted according to the statistical significance indicated by the p-values, and the top significant TFs can be prioritized as master regulators.

Master Regulator Analysis