kfwins2022 / epi-gene

Epi-gene

Epi Gene can cluster the whole genomes into 3 categories of core, accessory and unique genes. It can build two types of panmatrixes (i) Binary Panmatrix based on absence and presence of genes. (ii) Panmatrix based on the identity of the sequences. Binary pan-matrix will be used to build a UPMGA tree and heatmap to further describe an evolutionary relationship. While the pan-matrix based on sequence identity can be used for further quantitative analyses. This software can only work in WINDOWS operating systems. For a step by step guide, you can go through the vignettes, case study and this read me file.

               Install from GitHub :  devtools::install_github("furqan915/Epi-Gene")

Requirements:

Windows operating system

R-Language

R-Studio (Recommended).

Usearch software (available at www.drive5.com/usearch)

Prodigal software (http://compbio.ornl.gov/prodigal/)

Getting started

getwd()

setwd("E:/test")

library(EpiGene)

Prepare and load the meta-data file

genome.table <- read.table("E:/test/Genome.txt", sep="\t", header=TRUE)

genome.table

Predict the ORF

for( i in 1:dim(genome.table)[1] ){

cat("Relabelling the fasta sequences", genome.table$Files[i], "...\n")

in.file <- file.path("E:/test/genomes", genome.table$Files[i])

out.file <- file.path("E:/test/predicted", genome.table$Files[i])

predORFaa(in.file, out.file)}

Relabel the genome sequences

for( i in 1:dim(genome.table)[1] ){

cat("Relabelling the fasta sequences", genome.table$Files[i], "...\n")

in.file <- file.path("E:/test/genomes", genome.table$Files[i])

label <- file.path(genome.table$Relabel_ID[i])

out.file <- file.path("E:/test/relabel", genome.table$Files[i])

relabel(in.file,label,out.file)}

Join the relabelled genome sequences

setwd("E:/test/relabel")

joinfasta(allfasta)

Sorting by Length

Need to copy that combined.fasta file into the same very folder that have usearch.exe

sortbylength("combined.fasta", "sorted.fasta")

Clustering that returns the Binary Matrix

clust_bin("sorted.fasta", 0.5, "clusterd.fasta", 14)

Enumerating the genes in their related categories

panGen("bin_matrix.csv")

coregenes("bin_matrix.csv")

accessorygenes("bin_matrix.csv")

Uniqgenes("bin_matrix.csv")

Representative gene cluster sequences

core_gen("clusterd.fasta", 14, "coregen.fasta")

accessory_gen("clusterd.fasta", 14, "accessory.fasta")

uniq_gen("clusterd.fasta", 1, "uniq.fasta")

Phylogenetic Analyses

Dendrogram generation and clustered distances

distGen("bin_matrix.csv")

Heat Map chart based on the clustered distances

heatgen1("bin_matrix.csv")

Heat Map chart based on the presence/absence of genes

heatgen2("bin_matrix.csv")

This function needs heavy computation. Therefore high number of genomes requires a powerfull computer to generate this heat map.

Clustering on the basis of sequence Identity percentage

clust_id("sorted.fasta", 0.5, "clusterd_id.fasta")

Identity matrix generated after this clustering can be utilized for further quantitative analyses e.g. Prinicpal component Analyses.