Permalink
Cannot retrieve contributors at this time
Name already in use
A tag already exists with the provided branch name. Many Git commands accept both tag and branch names, so creating this branch may cause unexpected behavior. Are you sure you want to create this branch?
master_project_JLU2018/bin/cdhit_wrapper.R
Go to fileThis commit does not belong to any branch on this repository, and may belong to a fork outside of the repository.
102 lines (88 sloc)
5.1 KB
This file contains bidirectional Unicode text that may be interpreted or compiled differently than what appears below. To review, open the file in an editor that reveals hidden Unicode characters.
Learn more about bidirectional Unicode characters
#! /bin/Rscript | |
library("optparse") | |
option_list <- list( | |
make_option(opt_str = c("-i", "--input"), default = NULL, help = "Input bed-file. Fourth column is expected to contain names, last column must be sequences.", metavar = "character"), | |
make_option(opt_str = c("-c", "--identity"), default = 0.8, help = "Identity threshold. Default = %default", metavar = "double >= 0.8"), | |
make_option(opt_str = c("-A", "--coverage"), default = 8, help = "Minimal alignment length for both sequences in nucelotides. Default = %default", metavar = "integer"), | |
make_option(opt_str = c("-o", "--output"), default = "cluster.bed", help = "Output file same as input but with appended column of cluster numbers. Default = %default", metavar = "character"), | |
make_option(opt_str = c("--clean"), default = TRUE, help = "Delete all temporary files. Default = %default", metavar = "logical"), | |
# make_option(opt_str = c("-G", "--global"), default = FALSE), | |
# make_option(opt_str = c("-b", "--band_width"), default = 20), | |
# make_option(opt_str = c("-M", "--memory"), default = 800), | |
make_option(opt_str = c("-T", "--threads"), default = 1)#, | |
# make_option(opt_str = c("-n", "--word_length"), default = 3), | |
# make_option(opt_str = c("-l", "--throw_away_sequences"), default = 5), | |
# make_option(opt_str = c("-d", "--description")), # can not produce bed if this is != 0 | |
# make_option(opt_str = c("-s", "--length_dif_cutoff_shorter_p"), default = 0.0), | |
# make_option(opt_str = c("-S", "--length_dif_cutoff_shorter_n"), default = 999999), | |
# make_option(opt_str = c("-aL", "--alignment_coverage_longer_p"), default = 0.0), | |
# make_option(opt_str = c("-AL", "--alignment_coverage_longer_n"), default = 99999999), | |
# make_option(opt_str = c("-aS", "--alignment_coverage_shorter_p"), default = 0.0), | |
# make_option(opt_str = c("-AS", "--alignment_coverage_shorter_n"), default = 99999999), | |
# make_option(opt_str = c("-uL", "--max_unmatched_longer_p"), default = 1.0), | |
# make_option(opt_str = c("-uS", "--max_unmatched_shorter_p"), default = 99999999), | |
# make_option(opt_str = c("-U", "--max_unmatched_both_n"), default = 99999999), | |
# make_option(opt_str = c("-g", "--fast_cluster"), default = FALSE), | |
# make_option(opt_str = c("-r", "--ignore_strand"), default = TRUE), | |
# make_option(opt_str = c("--match"), default = 2), | |
# make_option(opt_str = c("--mismatch"), default = -2), | |
# make_option(opt_str = c("--gap"), default = -6), | |
# make_option(opt_str = c("--gap_ext"), default = -1), | |
# make_option(opt_str = c("-sc", "--sort_cluster_by_size"), default = TRUE) | |
# TODO more args | |
) | |
opt_parser <- OptionParser(option_list = option_list, | |
description = "CD-HIT-EST Wrapper function.") | |
opt <- parse_args(opt_parser) | |
#' cd-hit wrapper | |
#' | |
#' @param input Data.table or file in bed format (requires names in fourth column and sequences in last column). | |
#' @param identity Identity threshold. | |
#' @param coverage Minimal alignment length for both sequences in nucelotides. | |
#' @param output Clustered bedfile. Adds cluster number in last column (lower number = bigger cluster). | |
#' @param clean Clean up after run. | |
#' @param threads Number of threads to use in cd-hit (0 = all cpus). | |
#' | |
#' TODO add all cdhit parameter (necessary ?) | |
#' TODO check whether cdhit is installed | |
cdhitest <- function(input, identity = 0.8, coverage = 8, output = "cluster.bed", clean = TRUE, threads = 1) { | |
message("Loading bed.") | |
# load bed if neccessary | |
if (!data.table::is.data.table(input)) { | |
bed_table <- data.table::fread(input = input, header = FALSE) | |
} else { | |
bed_table <- input | |
} | |
message("Convert to fasta.") | |
### convert bed to fasta | |
# 4th column = name | |
# last column = sequence | |
fasta_file <- "converted_bed.fasta" | |
seqinr::write.fasta(sequences = as.list(bed_table[[ncol(bed_table)]]), names = bed_table[[4]], as.string = TRUE, file.out = fasta_file) | |
message("Clustering.") | |
### cd-hit-est | |
cdhit_output <- "cdhit_output" | |
cdhit_call <- paste("cd-hit-est -i", fasta_file, "-o", cdhit_output, "-c", identity, "-A", coverage, "-T ", threads, "-G 0 -n 3 -g 1 -r 0 -l 5 -sc 1 -d 0") | |
system(command = cdhit_call, wait = TRUE) | |
message("Formatting/ writing results.") | |
# reformat cluster file | |
# columns: id, clstr, clstr_size, length, clstr_rep, clstr_iden, clstr_cov | |
cluster_file <- "reformated.clstr" | |
cluster_call <- paste("clstr2txt.pl", paste0(cdhit_output, ".clstr"), ">", cluster_file) | |
system(command = cluster_call, wait = TRUE) | |
# load reformated file | |
cluster <- data.table::fread(cluster_file) | |
### add cluster to bed_table | |
result <- merge(x = bed_table, y = cluster[, c("id", "clstr")], by.x = "V4", by.y = "id", sort = FALSE)[, union(names(bed_table), names(cluster)[2]), with = FALSE] | |
# delete files | |
if (clean) { | |
file.remove(fasta_file, paste0(cdhit_output, ".clstr"), cdhit_output, cluster_file) | |
} | |
data.table::fwrite(x = result, file = output, sep = "\t", col.names = FALSE) | |
} | |
# call function with given parameter if not in interactive context (e.g. run from shell) | |
if (!interactive()) { | |
# remove last parameter (help param) | |
params <- opt[-length(opt)] | |
do.call(cdhitest, args = params) | |
} |