remove hardcoded path from cobbler/rails

Author: warrenlr

Dockerfile

@@ -16,9 +16,9 @@ RUN apt-get update -qq && apt-get install -y bwa wget cpanminus
 #
 # Download the software
 #
-RUN wget https://github.com/warrenlr/RAILS/raw/master/rails_v1-1.tar.gz && tar xvfz rails_v1-1.tar.gz && rm rails_v1-1.tar.gz 
+RUN wget https://github.com/bcgsc/RAILS/tree/master/tarball/rails_v1-4-1.tar.gz && tar xvfz rails_v1-4-1.tar.gz && rm rails_v1-4-1.tar.gz 
 
 #
 # Set the default working directory
 #
-WORKDIR /RAILS_v1.1
+WORKDIR /RAILS_v1.4.1

bin/RAILS

@@ -23,20 +23,21 @@
 use strict;
 use Getopt::Std;
 use Net::SMTP;
-use vars qw($opt_f $opt_s $opt_d $opt_i $opt_e $opt_l $opt_a $opt_v $opt_b $opt_t $opt_p $opt_q $opt_g);
-getopts('f:s:d:e:l:a:v:b:t:p:i:q:g:');
+use vars qw($opt_f $opt_s $opt_d $opt_i $opt_e $opt_l $opt_a $opt_v $opt_b $opt_t $opt_p $opt_q $opt_g $opt_p);
+getopts('f:s:d:e:l:a:v:b:t:p:i:q:g:p:');
 my ($base_name,$anchor,$seqid,$insert_stdev,$min_links,$max_link_ratio,$grace,$verbose)=("",1000,0.9,1.0,1,0.99,1,0,);
 
-my $version = "[v1.4]";
+my $version = "[v1.4.1]";
 my $dev = "rwarren\@bcgsc.ca";
-my $SAMPATH = "/gsc/btl/linuxbrew/bin/samtools";
+my $SAMPATH = "";
 #-------------------------------------------------
 
-if(! $opt_f || ! $opt_s || ! $opt_q){
+if(! $opt_f || ! $opt_s || ! $opt_q || ! $opt_p){
    print "Usage: $0 $version\n";
    print "-f  Assembled Sequences to further scaffold (Multi-Fasta format, required)\n"; 
    print "-q  File of filenames containing long Sequences queried (Multi-Fasta format, required)\n";
    print "-s  File of filenames containing full path to BAM file(s)\n";
+   print "-p  Full path to samtools (known to work/tested with v1.8, required)\n";
    print "-d  Anchoring bases on contig edges (ie. minimum required alignment size on contigs, default -d $anchor, optional)\n";
    print "-i  Minimum sequence identity fraction (0 to 1), default -i $seqid, optional\n";
    print "-t  LIST of names/header, long sequences to avoid using for merging/gap-filling scaffolds (optional)\n"; 
@@ -62,6 +63,7 @@ $max_link_ratio = $opt_a if($opt_a);
 my $listfile = $opt_t if($opt_t);
 $base_name = $opt_b if($opt_b);
 $grace = $opt_g if($opt_g);
+$SAMPATH = $opt_p if($opt_p);
 
 my $assemblyruninfo="";
 
@@ -73,6 +75,9 @@ if(! -e $fof){
    die "Invalid file: $fof -- fatal\n";
 }
 
+if(! -e $SAMPATH){
+   die "Invalid: $SAMPATH -- fatal\n";
+}
 
 ### Naming output files
 if ($base_name eq ""){

bin/cobbler.pl

@@ -23,21 +23,22 @@
 use strict;
 use Getopt::Std;
 use Net::SMTP;
-use vars qw($opt_f $opt_s $opt_d $opt_i $opt_v $opt_b $opt_t $opt_q $opt_l $opt_g);
-getopts('f:s:d:v:b:t:i:q:g:l:');
+use vars qw($opt_f $opt_s $opt_d $opt_i $opt_v $opt_b $opt_t $opt_q $opt_l $opt_g $opt_p);
+getopts('f:s:d:v:b:t:i:q:g:l:p:');
 my ($base_name,$anchor,$seqid,$verbose,$minreads,$grace)=("",1000,0.9,0,1,1);
 
-my $version = "[v0.5]";
+my $version = "[v0.5.1]";
 my $dev = "rwarren\@bcgsc.ca";
-my $SAMPATH = "/gsc/btl/linuxbrew/bin/samtools";
+my $SAMPATH = "";
 
 #-------------------------------------------------
 
-if(! $opt_f || ! $opt_s || ! $opt_q){
+if(! $opt_f || ! $opt_s || ! $opt_q || ! $opt_p){
    print "Usage: $0 $version\n";
    print "-f  Assembled Sequences to further scaffold (Multi-FASTA format NO LINE BREAKS, required)\n"; 
    print "-q  File of filenames containing long Sequences queried (Multi-FASTA format NO LINE BREAKS, required)\n";
    print "-s  File of filenames containing full path to BAM file(s) (use v0.2 for reading SAM files)\n";
+   print "-p  Full path to samtools (known to work/tested with v1.8, required)\n";
    print "-d  Anchoring bases on contig edges (ie. minimum required alignment size on contigs, default -d $anchor, optional)\n";
    print "-i  Minimum sequence identity fraction (0 to 1), default -i $seqid, optional\n";
    print "-l  Minimum number of long sequence support per gap, default -l $minreads, optional\n";
@@ -58,6 +59,7 @@
 $base_name = $opt_b if($opt_b);
 $grace = $opt_g if($opt_g);
 $minreads = $opt_l if($opt_l);
+$SAMPATH = $opt_p if($opt_p);
 
 my $assemblyruninfo="";
 
@@ -68,7 +70,9 @@
 if(! -e $fof){
    die "Invalid file: $fof -- fatal\n";
 }
-
+if(! -e $SAMPATH){
+   die "Invalid : $SAMPATH -- fatal\n";
+}
 
 ### Naming output files
 if ($base_name eq ""){

bin/runRAILS.sh

@@ -1,7 +1,7 @@
 #!/bin/bash
 #RLW 2016
-if [ $# -ne 4 ]; then
-        echo "Usage: $(basename $0) <FASTA assembly .fa> <FASTA long sequences .fa> <anchoring sequence length eg. 250> <min sequence identity 0.95>"
+if [ $# -ne 5 ]; then
+        echo "Usage: $(basename $0) <FASTA assembly .fa> <FASTA long sequences .fa> <anchoring sequence length eg. 250> <min sequence identity 0.95> <path to samtools>"
         exit 1
 fi
 ###Change line below to point to path of bwa executables
@@ -21,7 +21,7 @@ bwa mem -a -t4 $1-formatted.fa $2-formatted.fa | samtools view -Sb - > $2_vs_$1_
 echo Scaffolding $1-formatted.fa using $2-formatted.fa and filling gaps with sequences in $2-formatted.fa
 echo $2-formatted.fa > $2-formatted.fof
 echo $2_vs_$1_gapfilling.bam > $2_vs_$1_gapfilling.fof
-../cobbler.pl -f $1 -s $2_vs_$1_gapfilling.fof -d $3 -i $4 -b $2_vs_$1_$3_$4_gapsFill -q $2-formatted.fof
+../cobbler.pl -f $1 -s $2_vs_$1_gapfilling.fof -d $3 -i $4 -b $2_vs_$1_$3_$4_gapsFill -q $2-formatted.fof -p $5
 echo Process terminated.
 echo RAILS scaffolding $1.gapsFill.fa sequences using long seqs $2 -- anchoring sequence threshold $3 bp 
 echo reformatting file $1.gapsFill.fa
@@ -34,5 +34,5 @@ bwa mem -a -t4 $2_vs_$1_$3_$4_gapsFill-formatted.fa $2-formatted.fa | samtools v
 echo Scaffolding $2_vs_$1_$3_$4_gapsFill-formatted.fa using $2-formatted.fa and filling new gaps with sequences in $2-formatted.fa
 echo $2-formatted.fa > $2-formatted.fof
 echo $2_vs_$1_scaffolding.bam > $2_vs_$1_scaffolding.fof
-../RAILS -f $2_vs_$1_$3_$4_gapsFill-formatted.fa -s $2_vs_$1_scaffolding.fof -d $3 -i $4 -b $2_vs_$1_$3_$4_rails -q $2-formatted.fof
+../RAILS -f $2_vs_$1_$3_$4_gapsFill-formatted.fa -s $2_vs_$1_scaffolding.fof -d $3 -i $4 -b $2_vs_$1_$3_$4_rails -q $2-formatted.fof -p $5
 echo RAILS process terminated.

readme.md

@@ -1,14 +1,11 @@
-![Logo](https://github.com/warrenlr/rails/blob/master/rails-logo.png)
-
-# RAILS v1.4 and Cobbler v0.5
+# RAILS v1.4.1 and Cobbler v0.5.1
 ## Rene L. Warren, 2014-2018
 ## email: rwarren at bcgsc.ca
 
 ### Name
 -------------
 
 RAILS: Radial Assembly Improvement by Long Sequence Scaffolding
-
 Cobbler: Gap-filling with long sequences
 
 
@@ -32,6 +29,8 @@ You can test the software by executing "runme.sh" in the test folder. A simulate
 -------------
 
 RAILS and Cobbler are implemented in PERL and run on any OS where PERL is installed.
+Both tools require samtools (tested with v1.8) to read sequence alignment bamfiles. 
+The runRAILS.sh pipeline requires bwa (see Dependencies below for tested version).
 
 
 ### Community guidelines:
@@ -46,8 +45,8 @@ I encourage the community to contribute to the development of this software, by
 Download the tar ball, gunzip and extract the files on your system using:
 
 <pre>
-gunzip rails_v1-4.tar.gz
-tar -xvf rails_v1-4.tar
+gunzip rails_v1-4-1.tar.gz
+tar -xvf rails_v1-4-1.tar
 </pre>
 
 Alternatively, individual tools are available within the github repository
@@ -57,6 +56,8 @@ Alternatively, individual tools are available within the github repository
 -------------
 
 Make sure you have installed bwa (Version: 0.7.15-r1140) and that is is in your path.
+Make sure you have installed samtools (Version: 1.8) and that is is in your path.
+Compatible tools may be used, but have not been tested fully (eg. minimap2, sambamba)
 
 
 ### Test data
@@ -66,6 +67,8 @@ Make sure you have installed bwa (Version: 0.7.15-r1140) and that is is in your
 Go to ./test
 (cd test)
 
+You may need to change both runme.sh and runmeHuman.sh to specify the path of samtools on your system
+
 1. SARS:
 execute runme.sh
 (./runme.sh)
@@ -94,7 +97,7 @@ Software. doi: 10.21105/joss.00116
 
 <pre>
 ./runRAILS.sh
-Usage: runRAILS.sh <FASTA assembly .fa> <FASTA long sequences .fa> <anchoring sequence length eg. 250> <min sequence identity 0.95>
+Usage: runRAILS.sh <FASTA assembly .fa> <FASTA long sequences .fa> <anchoring sequence length eg. 250> <min sequence identity 0.95> <path to samtools>
 
 this pipeline will:
 1. reformat the assembly file $1
@@ -107,10 +110,11 @@ this pipeline will:
 8. Align the reformatted long sequences to your re-formatted cobbler assembly
 9. Run RAILS to generate a newly scaffolded assembly draft
 
-Usage: ./cobbler.pl [v0.5]
+Usage: ./cobbler.pl [v0.5.1]
 -f  Assembled Sequences to further scaffold (Multi-FASTA format NO LINE BREAKS, required)
 -q  File of filenames containing long Sequences queried (Multi-FASTA format NO LINE BREAKS, required)
 -s  File of filenames containing full path to BAM file(s) (use v0.2 for reading SAM files)
+-p  Full path to samtools (known to work/tested with v1.8, required)
 -d  Anchoring bases on contig edges (ie. minimum required alignment size on contigs, default -d 1000, optional)
 -i  Minimum sequence identity fraction (0 to 1), default -i 0.9, optional
 -l  Minimum number of long sequence support per gap, default -l 1, optional
@@ -121,10 +125,11 @@ Usage: ./cobbler.pl [v0.5]
 IMPORTANT: the order of files in -q and -s MUST match!
 
 
-Usage: ./RAILS [v1.4]
+Usage: ./RAILS [v1.4.1]
 -f  Assembled Sequences to further scaffold (Multi-Fasta format, required)
 -q  File of filenames containing long Sequences queried (Multi-Fasta format, required)
 -s  File of filenames containing full path to BAM file(s)
+-p  Full path to samtools (known to work/tested with v1.8, required)
 -d  Anchoring bases on contig edges (ie. minimum required alignment size on contigs, default -d 1000, optional)
 -i  Minimum sequence identity fraction (0 to 1), default -i 0.9, optional
 -t  LIST of names/header, long sequences to avoid using for merging/gap-filling scaffolds (optional)