miga-base 0.7.26.0 → 1.0.0.1

data/utils/enveomics/Scripts/Newick.autoprune.R

@@ -0,0 +1,27 @@
+#!/usr/bin/env Rscript
+#
+# @author  Luis M. Rodriguez-R
+# @update  Jan-07-2016
+# @license artistic license 2.0
+#
+
+#= Load stuff
+args <- commandArgs(trailingOnly = F)  
+enveomics_R <- file.path(dirname(
+   sub("^--file=", "", args[grep("^--file=", args)])),
+   "lib", "enveomics.R")
+source(file.path(enveomics_R, "R", "cliopts.R"))
+source(file.path(enveomics_R, "R", "autoprune.R"))
+
+#= Generate interface
+opt <- enve.cliopts(enve.prune.dist,
+   file.path(enveomics_R, "man", "enve.prune.dist.Rd"),
+   positional_arguments=1,
+   usage="usage: %prog [options] output.nwk",
+   mandatory=c("t"),
+   number=c("min_dist","order"),
+   o_desc=list(t="A tree to prune in Newick format."))
+
+#= Run it!
+pt <- do.call("enve.prune.dist", opt$options)
+ape::write.tree(pt, opt$args[1])

data/utils/enveomics/Scripts/RAxML-EPA.to_iToL.pl

@@ -0,0 +1,228 @@
+#!/usr/bin/env perl
+
+# @author Luis M Rodriguez-R <lmrodriguezr at gmail dot com>
+# @update Mar-23-2015
+# @license Artistic License 2.0
+
+use warnings;
+use strict;
+use List::Util qw/sum max/;
+use Getopt::Std;
+use Math::Round qw/round/;
+our $VERSION = 1.1;
+
+warn <<WARN
+
+
+┌──[ IMPORTANT ]─────────────────────────────────────────────────┐
+│ This script has been deprecated in favor of JPlace.to_iToL.rb. │
+│ Please use the new version, together with the RAxML EPA's file │
+│ RAxML_portableTree.*.jplace instead.                           │
+└────────────────────────────────────────────────────────────────┘
+WARN
+;
+sub HELP_MESSAGE {
+die "
+Description:
+   Reformats the node names (labels) of a RAxML_originalLabelledTree.<NAME> file
+   (produced by RAxML's EPA, -f v), so it can be opened in most tree viewers (like
+   iToL and FigTree).  Also, it creates iToL-compatible files to draw pie-charts
+   (based on the classification of short reads) in the nodes of the reference tree.
+
+Usage:
+   $0 -n <NAME> [other options...]
+
+   -n <str> *	Name of the run used in RAxML.
+   -t <str>	Use this file as original labelled tree, instead of generating one
+   		based on the job name.  By default, RAxML_originalLabelledTree.<NAME>
+		in the -d directory. See [NOTE1].
+   -d <str>	Directory containing RAxML files.  By default: current directory.
+   -o <str>	Output tree.  By default, it takes the path to the input tree and
+   		appends .nwk to it.
+   -l <str>	File containing a list of internal nodes.  The nodes in the list
+   		will be renamed, and the reads of all children nodes will be
+		transferred to it.  This can be useful if you want to display
+		these nodes collapsed.  The format of the file is raw text, with
+		two columns separated by tabs or spaces, where the first column is
+		the original name of the internal node (without the brackets) and
+		the second is the name to be used.  See [NOTE2].
+   -a		Append original label to the renamed nodes (only if -l is passed).
+   -s <str>	The names of the reads will be assumed to contain the sample name,
+   		separated by this string.  For example, if the value is '_', and
+		a read has the name 'hco_ABCDEF/1#ACTG', it will be assumed to be
+		a read from the sample 'hco'.  If not provided, all the reads are
+		assumed to come from the same sample (called 'unknown').
+   -m <str>	Comma-delimited list of samples.  If not provided, all found samples
+   		will be used (unsorted).
+   -c <str>	Comma-delimited list of colors (in RGB hexadecimal) to represent
+   		the different samples.  If not provided (or if insufficient values
+		are provided) random colors are generated.
+   -N <str>	Comma-delimited list of normalizing factors per dataset.  Typically,
+   		the size of the datasets divided by a fixed value (e.g. size x 1,000,
+		to express sizes as reads per thousand).
+   -T		Use the total number of assigned reads per sample (times a constant)
+   		as the normalizing factor. The constant used corresponds to the 100
+		times the size of the largest factor. If passed, -N is ignored.
+   -q		Run quietly.
+   -h/--help	Displays this message and exits.
+
+   * Mandatory
+   [NOTE1] The tree provided by -t MUST be based on a tree produced by this script
+   without the -l option.
+   [NOTE2] The tree produced by RAxML-EPA is usually not correctly rooted, which
+   makes the -l option useless.  However, you can manually root the tree and provide
+   the rooted tree in Newick format using the -t option.  If you do this, make
+   sure the program doesn't change/delete the names of the internal nodes.  I know
+   that iToL can do it correctly (if you export preserving the original IDs), while
+   FigTree deletes the labels.  I didn't try any other tool.
+
+";
+}
+
+my %o;
+getopts('n:t:d:o:l:s:m:c:N:Tqh', \%o);
+$o{d} ||= '.';
+$o{n} or &HELP_MESSAGE;
+$o{h} and &HELP_MESSAGE;
+$o{c} = [split /,/, (defined $o{c}?$o{c}:"")];
+$o{N} = [split /,/, (defined $o{N}?$o{N}:"")];
+
+# Set files
+my $inTree   = ($o{t} || $o{d}."/RAxML_originalLabelledTree.".$o{n});
+my $outTree  = ($o{o} || $inTree.".nwk");
+my $inClass  = $o{d}."/RAxML_classification.".$o{n};
+my $outClass = $inClass.".iToL";
+my $outColl  = $outTree.".collapse.iToL";
+
+# Relocate tree node names
+print STDERR "o Reformatting tree.\n" unless $o{q};
+open INTREE, "<", $inTree or die "Cannot read file: $inTree: $!\n";
+my $tree = <INTREE>;
+$tree =~ s/:([\d\.]+)(\[.+?\])/$2:$1/g unless $o{t};
+close INTREE;
+
+# Read leaf nodes
+print STDERR "o Reading nodes.\n" unless $o{q};
+my %tags    = ();
+
+my $t = $tree;
+while($t =~ m/([A-Za-z0-9_\|\.-]+\[([A-Za-z0-9_\|\.-]+)\])/){
+   my $n = $1;
+   my $ta = $2;
+   $tags{$ta} = $n;
+   $t = substr $t, (length($n) + index $t, $n);
+}
+
+# Label/collapse internal nodes
+if($o{l}){
+   print STDERR "o Labeling/collapsing internal nodes.\n";
+   open LIST, "<", $o{l} or die "Cannot read file: $o{l}: $!\n";
+   open COLL, ">", $outColl or die "Cannot create file: $outColl: $!\n";
+   while(<LIST>){
+      chomp;
+      next if /^#/ or /^\s*$/;
+      # Label internal node
+      my @l = split /\s+/;
+      $l[0] =~ m/^\[(.+)\]$/ or die "Unable to parse internal node name: $l[0].\n";
+      my $ori = $1;
+      my $new = $l[1];
+      if(exists $tags{$ori}){
+	 warn "Warning: Trying to label/collapse $ori as $new, already defined as $tags{$ori}.\n";
+	 next;
+      }
+      $new =~ s/[^A-Za-z0-9_\|\.\-]/_/g;
+      $new.= "[$ori]" if $o{a};
+      $tags{$ori} = $new;
+      $tree =~ s/\[$ori\]/$new/;
+      # Isolate node
+      $t = substr $tree, 0, index($tree, $new);
+      my $i=length($t)-2;
+      for(my $c=1 ; $i and $c; $i--){
+         my $char = substr $t, $i, 1;
+	 $c++ if $char eq ')';
+	 $c-- if $char eq '(';
+      }
+      $t = substr $t, $i;
+      # Get children
+      $t =~ s/:[\d\.]+|[\(\)]/,/g;
+      $t =~ s/,+/,/g;
+      my $chn=0;
+      for my $child (split /,/, $t){
+         next unless $child;
+	 $child =~ s/.*\[(.+?)\]/$1/;
+	 $tags{$child} = $new;
+	 $chn++;
+      }
+      print STDERR "  Collapsing $new: $chn children.\n" unless $o{q};
+      print COLL "$new\n";
+   }
+   close LIST;
+   close COLL;
+}
+
+# Save tree
+open OUTTREE, ">", $outTree or die "Cannot create file: $outTree: $!\n";
+print OUTTREE $tree;
+close OUTTREE;
+
+# Count reads
+my %samples = ();
+my %nodes   = ();
+print STDERR "o Counting reads.\n";
+my $s = defined $o{s} ? $o{s} : "";
+open INCLASS, "<", $inClass or die "Cannot read file: $inClass: $!\n";
+while(<INCLASS>){
+   my @ln = split /\s+/;
+   $ln[0] =~ s/$s.+$//; # Sample name
+   ($samples{$ln[0]} ||= 0)++;
+   $tags{$ln[1]} ||= "[".$ln[1]."]"; # Node name
+   (($nodes{$tags{$ln[1]}} ||= {})->{$ln[0]} ||= 0)++;
+}
+close INCLASS;
+
+
+my $labs = 'LABELS';
+my $cols = 'COLORS';
+my @samples = $o{m} ? (split /,/, $o{m}) : (keys %samples);
+my @normfac = ();
+for my $sample (@samples){
+   my $col = shift @{$o{c}};
+   unless(defined $col and length($col)==6){
+      $col = '';
+      for (1 .. 3){
+	 my $v = int rand 16;
+	 $v = chr $v+55 if $v>9;
+	 $col.="$v$v";
+      }
+   }
+   my $nf = shift @{$o{N}};
+   $nf = 1 unless defined $nf and $nf>0;
+   $labs.= ','.($sample || 'unknown');
+   $cols.= ',#'.$col;
+   push @normfac, $nf+0;
+}
+
+open OUTCLASS, ">", $outClass or die "Cannot create file: $outClass: $!\n";
+print OUTCLASS "$labs\n$cols\n";
+my $tiny=0;
+for my $node (keys %nodes){
+   my $i=0;
+   for my $s (@samples){
+      $nodes{$node}->{$s} = ($nodes{$node}->{$s} || 0)/($o{T} ? ($samples{$s}||1)/(max(values %samples)*100) : ($normfac[$i++]||1));
+   }
+   my $r = round(sum(values %{$nodes{$node}}));
+   print OUTCLASS "$node,R$r";
+   for my $sample (@samples){
+      print OUTCLASS ",".round($nodes{$node}->{$sample} || 0);
+   }
+   print OUTCLASS "\n";
+   $tiny++ unless $r;
+}
+close OUTCLASS;
+
+unless($o{q}) {
+   print "Total counts per dataset:\n";
+   print "  $_\t".($samples{$_}||0)."\n" for @samples;
+}
+warn "$tiny node assignments are too small to represent. Decrease the values of -N or use an alternative like -T." if $tiny;
+

data/utils/enveomics/Scripts/RecPlot2.compareIdentities.R

@@ -0,0 +1,32 @@
+#!/usr/bin/env Rscript
+#
+# @author  Luis M. Rodriguez-R
+# @update  Jan-04-2016
+# @license artistic license 2.0
+#
+
+#= Load stuff
+args <- commandArgs(trailingOnly = F)  
+enveomics_R <- file.path(dirname(
+   sub("^--file=", "", args[grep("^--file=", args)])),
+   "lib", "enveomics.R")
+library(methods)
+source(file.path(enveomics_R, "R", "cliopts.R"))
+source(file.path(enveomics_R, "R", "recplot2.R"))
+
+#= Generate interface
+opt <- enve.cliopts(enve.recplot2.compareIdentities,
+   file.path(enveomics_R, "man", "enve.recplot2.compareIdentities.Rd"),
+   positional_arguments=2,
+   usage="usage: %prog [options] recplot-A.Rdata recplot-B.Rdata",
+   number=c("pseudocounts", "max.deviation"), ignore=c("x", "y"),
+   p_desc="Calculates the difference between identity distributions of two recruitment plots.")
+
+#= Run it!
+load(opt$args[1])
+opt$options[['x']] <- rp
+load(opt$args[2])
+opt$options[['y']] <- rp
+dist <- do.call("enve.recplot2.compareIdentities", opt$options)
+cat(dist, '\n')
+

data/utils/enveomics/Scripts/RefSeq.download.bash

@@ -0,0 +1,48 @@
+#!/bin/bash
+
+#
+# @author  Luis M. Rodriguez-R
+# @update  Oct-20-2015
+# @license artistic license 2.0
+#
+
+FTP="ftp://ftp.ncbi.nlm.nih.gov/genomes/refseq/bacteria"
+ORG=$1
+EXT=${2:-.*.gz}
+STT=${3:-Any}
+DIR=${4:-$ORG}
+
+if [[ "$ORG" == "" ]] ; then
+echo "
+Downloads a collection of sequences and/or annotations from NCBI's RefSeq.
+
+Usage:
+$0 <organism> [<extension>[ <level>[ <dir>]]]
+
+<organism>	The organism to download (e.g., Streptococcus_pneumoniae).
+<extension>	Extension to download.  Common extensions include '.fna.gz'
+		(genomic sequences), '.faa.gz' (protein sequences), and
+		'.gff.gz' (annotations).  By default: '.*.gz' (all data).
+<level>		Use only genomes with this assembly level. Common levels are
+		'Complete Genome' and 'Contig'.  By default, any assembly
+		level is allowed ('Any').
+<dir>		Directory where the files are to be downladed. By default,
+		same as <organism>.
+" >&2
+exit
+fi
+
+[[ -d "$DIR" ]] || mkdir "$DIR"
+
+curl -s "$FTP/$ORG/assembly_summary.txt" -o "$DIR/assembly_summary.txt"
+for path in $(cat "$DIR/assembly_summary.txt" \
+      | awk -F"\t" "\$12==\"$STT\" || \"$STT\"==\"Any\" {print \$20}" ) ; do
+   dir="$DIR/$(basename "$path")"
+   [[ -d "$dir" ]] || mkdir "$dir"
+   for file in $(curl -s "$path/" | awk '{print $9}') ; do
+      if [[ "$file" == *$EXT ]] ; then
+	 curl -s "$path/$file" -o "$dir/$file"
+      fi
+   done
+done
+

data/utils/enveomics/Scripts/SRA.download.bash

@@ -0,0 +1,55 @@
+#!/bin/bash
+
+DATA_LINK="https://www.ebi.ac.uk/ena/portal/api/filereport"
+DATA_OPS="result=read_run&fields=run_accession,fastq_ftp,fastq_md5"
+SRX=$1
+DIR=${2:-$SRX}
+VERSION=1.0
+
+if [[ "$SRX" == "" ]] ; then
+echo "
+[Enveomics Collection: $(basename "$0" .bash) $VERSION]
+
+Downloads the set of runs from a project, sample, or experiment in SRA.
+
+Usage:
+$(basename "$0") <SRA-ID>[ <dir>]
+
+<SRA-ID>	ID of the SRA Project, Sample, or Experiment.
+<dir>		Directory where the files are to be downladed. By default,
+		same as <SRA-ID>.
+" >&2
+exit
+fi
+
+[[ -d "$DIR" ]] || mkdir "$DIR"
+
+function md5value {
+  local file=$1
+  o=$(md5 "$file" | perl -pe 's/.* //')
+  [[ -n $o ]] || o=$(md5sum-lite "$file" | awk '{print $1}')
+  [[ -n $o ]] || o=$(md5sum "$file" | awk '{print $1}')
+  echo "$o"
+}
+
+curl -Ls "$DATA_LINK?$DATA_OPS&accession=$SRX" -o "$DIR/srr_list.txt"
+tail -n +2 "$DIR/srr_list.txt" | while read ln ; do
+  srr=$(echo "$ln"|cut -f 1)
+  ftp=$(echo "$ln"|cut -f 2)
+  md5=$(echo "$ln"|cut -f 3)
+  dir="$DIR/$srr"
+  [[ -d "$dir" ]] || mkdir "$dir"
+  echo "o $srr" >&2
+  for uri in $(echo "$ftp" | tr ";" " ") ; do
+    file="$dir/$(basename $uri)"
+    curl "$uri" -o "$file"
+    md5obs=$(md5value "$file" 2> /dev/null)
+    if [[ "$md5" == "$md5obs"* ]] ; then
+      md5=$(echo "$md5" | perl -pe 's/^[^;]+;//')
+    else
+      echo "Corrupt file: $file" >&2
+      echo "  MD5 mismatch: $md5obs not in $md5" >&2
+      exit 1;
+    fi
+  done
+done

data/utils/enveomics/Scripts/TRIBS.plot-test.R

@@ -0,0 +1,36 @@
+#!/usr/bin/env Rscript
+#
+# @author  Luis M. Rodriguez-R
+# @update  Jan-05-2016
+# @license artistic license 2.0
+#
+
+#= Load stuff
+suppressPackageStartupMessages(library(enveomics.R))
+args <- commandArgs(trailingOnly = F)  
+enveomics_R <- file.path(dirname(
+   sub("^--file=", "", args[grep("^--file=", args)])),
+   "lib", "enveomics.R")
+
+#= Generate interface
+opt <- enve.cliopts(plot.enve.TRIBStest,
+   file.path(enveomics_R, "man", "plot.enve.TRIBStest.Rd"),
+   positional_arguments=c(1,3),
+   usage="usage: %prog [options] output.pdf [width height]",
+   mandatory=c("x"),
+   vectorize=c("xlim","ylim"),
+   number=c("xlim","ylim"),
+   defaults=c(type="overlap", xlim=NA, ylim=NA))
+
+#= Run it!
+a <- new.env()
+load(opt$options[['x']], a)
+opt$options[['x']] <- get(ls(envir=a),envir=a)
+summary(opt$options[['x']])
+if(is.na(opt$options[['xlim']][1])) opt$options[['xlim']] <- NULL
+if(is.na(opt$options[['ylim']][1])) opt$options[['ylim']] <- NULL
+args = as.list(opt$args)
+for(i in 2:3) if(length(args)>=i) args[[i]] <- as.numeric(args[[i]])
+do.call("pdf", args)
+do.call("plot.enve.TRIBStest", opt$options)
+dev.off()

data/utils/enveomics/Scripts/TRIBS.test.R

@@ -0,0 +1,39 @@
+#!/usr/bin/env Rscript
+#
+# @author  Luis M. Rodriguez-R
+# @update  Jan-05-2016
+# @license artistic license 2.0
+#
+
+#= Load stuff
+suppressPackageStartupMessages(library(enveomics.R))
+args <- commandArgs(trailingOnly = F)  
+enveomics_R <- file.path(dirname(
+   sub("^--file=", "", args[grep("^--file=", args)])),
+   "lib", "enveomics.R")
+
+#= Generate interface
+opt <- suppressWarnings(enve.cliopts(enve.tribs,
+   file.path(enveomics_R, "man", "enve.tribs.Rd"),
+   positional_arguments=c(0,2),
+   usage="usage: %prog [options] [output.Rdata [bins=50]]",
+   mandatory=c("dist", "selection"),
+   defaults=c(dimensions=0, selection=NULL),
+   ignore=c("metaMDS.opts","points","pre.tribs","subsamples"),
+   o_desc=list(dist="A tab-delimited matrix of distances.",
+      selection="A list of names with the selection to evaluate."),
+   p_desc=paste("",
+      "Estimates the empirical difference between all the distances",
+      "in a set of objects and a subset, together with its statistical",
+      "significance.",sep="\n\t")))
+
+#= Run it!
+opt$options[['dist']] <- as.dist(read.table(opt$options[['dist']],
+   header=TRUE, sep="\t", row.names=1))
+opt$options[['selection']] <- read.table(opt$options[['selection']],
+   header=FALSE, sep="\t", as.is=TRUE)[,1]
+if(opt$options[['dimensions']]==0) opt$options[['dimensions']] <- NULL
+if(length(opt$args)>1) opt$options[['bins']] <- as.numeric(opt$args[2])
+t <- do.call("enve.tribs.test", opt$options)
+summary(t)
+if(length(opt$args)>0) save(t, file=opt$args[1])