--- a/nanopore_qc.R	Mon Mar 12 19:55:54 2018 -0400
+++ b/nanopore_qc.R	Sun Aug 12 13:21:30 2018 -0400
@@ -56,11 +56,20 @@
                      help="The cutoff value for the mean Q score of a read (default 7). Used to create separate plots for reads above and below this threshold"
                      )
 
+parser <- add_option(parser, 
+                     opt_str = c("-d", "--discard_failed"), 
+                     type="logical", 
+                     default=FALSE,
+                     dest = 'filt.failed',
+                     help="Discard reads that failed Albacore filtering"
+                     )
+
 opt = parse_args(parser)
 
-input.file = opt$input.file
-output.dir = opt$output.dir
-q = opt$q
+input.file  = opt$input.file
+output.dir  = opt$output.dir
+filt.failed = opt$filt.failed
+q           = opt$q
 
 # this is how we label the reads at least as good as q
 q_title = paste("Q>=", q, sep="")
@@ -123,6 +132,7 @@
     # by default the lowest value is -Inf, i.e. includes all reads. The 
     # other value in min.q is set by the user at the command line
     d = read_tsv(filepath, col_types = cols_only(channel = 'i', 
+                                                passes_filtering = 'c',
                                                 num_events_template = 'i', 
                                                 sequence_length_template = 'i', 
                                                 mean_qscore_template = 'n',
@@ -146,6 +156,10 @@
 
     # ignore 0-length reads
     d <- d[d$sequence_length_template > 0,]
+    # ignore reads failing filtering
+    if (filt.failed) {
+        d <- d[d$passes_filtering == 'True',]
+    }
         
     d$events_per_base = d$num_events_template/d$sequence_length_template
 
@@ -173,7 +187,6 @@
     d = d[keep]
 
     d$start_bin = cut(d$start_time, 9,labels=c(1:9))
-    write.table(d,"foo.tsv",sep="\t",quote=F)
         
     return(d)
 }