add script and process to extract unmerged seqs from the rds file

- mimics logic from old NGS16S pipeline script as described in #17

bin/get_dropped_chim.R

@@ -0,0 +1,28 @@
+#!/usr/bin/env Rscript
+
+suppressPackageStartupMessages(library(argparse, quietly = TRUE))
+suppressPackageStartupMessages(library(tidyr, quietly = TRUE))
+suppressPackageStartupMessages(library(readr, quietly = TRUE))
+suppressPackageStartupMessages(library(dplyr, quietly = TRUE))
+n
+main <- function(arguments){
+  parser <- ArgumentParser(
+      description="Write forward and reverse unmerged denoised, dereplicated reads")
+  parser$add_argument('rdata', help='RDS files containing dada2 output')
+  parser$add_argument(
+             '-o', '--outfile', default='chim_dropped.csv',
+             help='output csv file with weight and sequence of svs identified as chimeras')
+
+  args <- parser$parse_args(arguments)
+
+  obj <- readRDS(args$rdata)
+  seqtab <- as.data.frame(as.table(obj$seqtab))
+  seqtab.nochim <- as.data.frame(as.table(obj$seqtab.nochim))
+
+  seqtab %>% anti_join(seqtab.nochim, by=c('Var2')) %>%
+        arrange(-Freq) %>% rename(sequence=Var2) %>%
+        rename(weight=Freq) %>% select(weight, sequence) %>%
+        write_csv(args$outfile)
+}
+
+main(commandArgs(trailingOnly=TRUE))

main.nf

@@ -308,6 +308,8 @@ process dada_get_unmerged {
 
     output:
         file("unmerged_*.fasta") into dada_unmerged
+        val sampleid into dada_unmerged_samples
+        file dada_rds into dada_unmerged_rds
 
     publishDir "${params.output}/dada/${sampleid}/", overwrite: true, mode: 'copy'
 
@@ -317,6 +319,26 @@ process dada_get_unmerged {
     """
 }
 
+process dada_get_dropped_chimeras {
+
+    label 'med_cpu_mem'
+
+    input:
+        val sampleid from dada_unmerged_samples
+        file("dada_params.json") from maybe_local(params.dada_params)
+        file dada_rds from dada_unmerged_rds
+
+    output:
+        file("chim_dropped.csv") into dada_chim_dropped
+
+    publishDir "${params.output}/dada/${sampleid}/", overwrite: true, mode: 'copy'
+
+    """
+    get_dropped_chim.R ${dada_rds} \
+        --outfile chim_dropped.csv
+    """
+}
+
 
 process combined_overlaps {