Summary tables for all samples

.gitignore

@@ -0,0 +1,4 @@
+.nextflow*
+out_*
+work
+.Rhistory

Dockerfile

@@ -1,11 +1,11 @@
 ################################
 #                              #
 #         DOCKERFILE           #
-#   Hybrid Assembly Pipeline   #
+#   plasmIDent Pipeline        #
 #                              #
 ################################
 FROM continuumio/miniconda3
-MAINTAINER Caspar Gross <[email protected]>
+LABEL author="Caspar Gross <[email protected]>"
 LABEL description="contains all the dependencies for plasmid Identification pipeline at github.com/caspargross/plasmidIdentification" 
 
 SHELL ["/bin/bash", "-c"]

bin/04_summary_table.R

@@ -46,7 +46,9 @@ getID <- function(s) {
   return(contig)
 }
 
-dt_ar <- dt_rgi[, .(contig = sapply(Contig, getID), ar_genes = sapply(Best_Hit_ARO, toString)),]
+dt_ar <- dt_rgi[, .(contig = sapply(Contig, getID), 
+                    ar_genes = sapply(Best_Hit_ARO, toString), 
+                    aro = sapply(ARO, toString)),]
 
 # Put it all together
 dt <- dt_gc[,.(gc = mean(gc)), by = contig]

bin/06_summary_samples.R

@@ -0,0 +1,42 @@
+#!/usr/bin/env Rscript
+
+########################
+# Create summary table #
+
+########################
+
+require(data.table)
+require(stringr)
+require(knitr)
+
+args <- commandArgs(trailingOnly=TRUE)
+
+if (length(args) >= 1) {
+	df.list <- lapply(args, fread)
+	names(df.list) <- str_remove(basename(args), "_summary.csv")
+
+	# tsv table
+	df <- data.table::rbindlist(df.list, idcol = "sample_name")
+	fwrite(df, sep = "\t", file = "rgi_summary_samples.tsv")
+
+	# html table, use aro_index.tsv from the card db to match aro accession to ID
+	#<a href="https://www.thesitewizard.com/" target="_blank">thesitewizard.com</a>
+	# https://card.mcmaster.ca/ontology/43314
+	aro <- fread("/aro_index.csv")
+	aro_dt <- data.table(aro = str_remove(aro$`ARO Accession`, "^ARO:"), 
+											id = aro$`CVTERM ID`)
+
+	df$aro <- as.character(df$aro)
+	dfm <- merge(df, aro_dt, by = "aro")
+	dfm$id <- paste0("<a href='https://card.mcmaster.ca/ontology/", dfm$id, "'", " target='_blank'>", dfm$id, "</a>")
+
+	cat(knitr::kable(dfm[, !c("aro")],
+									 format = "html", 
+									 escape = FALSE, 
+									 align = "r", 
+									 caption = "Summary table of all replicons with resistance genes"), 
+			file = "rgi_summary_samples.html")
+
+} else {
+	stop("No arguments provided")
+}

conf/circos/housekeeping.conf

@@ -218,7 +218,7 @@ offsets = 0,0
 max_ticks            = 5000
 max_ideograms        = 200
 max_links            = 25000
-max_points_per_track = 55000
+max_points_per_track = 255000
 
 # What to do when data is found for an ideogram that does not appear in the karyotype file.
 # Set to 'skip' or 'exit'

doc/alternative_installation.md

@@ -18,7 +18,7 @@ The conda environment containing all the dependencies for this pipeline can be i
 You need [miniconda](https://conda.io/miniconda.html) installed on you machine. You can then create the new environment with:
 
 ```
-conda create -f /env/PI_env.yml 
+conda env create -f env/PI_env.yml 
 ```
 
 Additionally you need to run the following commands:

main.nf

@@ -240,6 +240,7 @@ process format_data_rgi {
 
     output:
     set id, file("rgi.txt"), file("rgi_span.txt") into circos_data_rgi
+    //set id, file("rgi_summary.txt") into rgi_summary_ch
 
     script:
     """
@@ -302,6 +303,7 @@ process calcGC {
 // Calculate gc conten
     publishDir "${params.outDir}/${id}/gc", mode: 'copy'
     tag{id}
+    errorStrategy 'ignore'
 
     input:
     set id, assembly, lr, type from gc_padded
@@ -398,12 +400,13 @@ process table{
 // Create table with contig informations
     publishDir "${params.outDir}/${id}/", mode: 'copy'
     tag{id}
+    errorStrategy 'ignore'
 
     input:
     set id, gc, assembly, cov, type, rgi from table_data
 
     output:
-    file("${id}_summary.csv")
+    file("${id}_summary.csv") into summary_tsvs
 
     script:
     """
@@ -413,6 +416,30 @@ process table{
     """
 }
 
+aro_index = Channel.fromPath("${baseDir}/data/aro_index.csv")
+
+process summary_samples {
+// Create summary table with all samples
+publishDir "${params.outDir}", mode: 'copy'
+
+input: 
+path tsvs from summary_tsvs
+file 'aro_index'
+// aro_index.csv is in the docker image
+
+output:
+path "rgi_summary_samples.tsv"
+path "rgi_summary_samples.html"
+// out files produced by R script
+
+script:
+"""
+${env}
+06_summary_samples.R ${tsvs}
+"""
+
+}
+//summary_data.view()
 
 /*
 ================================================================================