miga-base 0.7.26.3 → 1.0.0.sr1

data/utils/enveomics/Scripts/sam.filter.rb

@@ -0,0 +1,148 @@
+#!/usr/bin/env ruby
+
+# frozen_string_literal: true
+
+$VERSION = 1.0
+$:.push File.expand_path('../lib', __FILE__)
+require 'enveomics_rb/enveomics'
+use 'shellwords'
+
+o = {
+  q: false, threads: 2, m_format: :sam, g_format: :fasta, identity: 95.0,
+  o: '-', header: true
+}
+
+OptionParser.new do |opt|
+  Enveomics.opt_banner(
+    opt, 'Filters a SAM or BAM file by target sequences and/or identity',
+    "#{File.basename($0)} -m map.sam -o filtered_map.sam [options]"
+  )
+
+  opt.separator 'Input/Output'
+  opt.on(
+    '-g', '--genome PATH',
+    'Genome assembly',
+    'Supports compression with .gz extension, use - for STDIN'
+  ) { |v| o[:g] = v }
+  opt.on(
+    '-m', '--mapping PATH',
+    'Mapping file',
+    'Supports compression with .gz extension, use - for STDIN'
+  ) { |v| o[:m] = v }
+  opt.on(
+    '-o', '--out-sam PATH',
+    'Output filtered file in SAM format',
+    'Supports compression with .gz extension, use - for STDOUT (default)'
+  ) { |v| o[:o] = v }
+  opt.separator ''
+
+  opt.separator 'Formats'
+  opt.on(
+    '--g-format STRING',
+    'Genome assembly format: fasta (default) or list'
+  ) { |v| o[:g_format] = v.downcase.to_sym }
+  opt.on(
+    '--m-format STRING',
+    'Mapping file format: sam (default) or bam',
+    'sam supports compression with .gz file extension'
+  ) { |v| o[:m_format] = v.downcase.to_sym }
+  opt.separator ''
+
+  opt.separator 'General'
+  opt.on(
+    '-i', '--identity FLOAT', Float,
+    "Set a fixed threshold of percent identity (default: #{o[:identity]})"
+  ) { |v| o[:identity] = v }
+  opt.on('--no-header', 'Do not include the headers') { |v| o[:header] = v }
+  opt.separator ''
+  opt.on(
+    '-t', '--threads INT', Integer, "Threads to use (default: #{o[:threads]})"
+  ) { |v| o[:threads] = v }
+  opt.on('-l', '--log PATH', 'Log file to save output') { |v| o[:log] = v }
+  opt.on('-q', '--quiet', 'Run quietly') { |v| o[:q] = v }
+  opt.on('-h', '--help', 'Display this screen') do
+    puts opt
+    exit
+  end
+  opt.separator ''
+end.parse!
+
+$QUIET = o[:q]
+
+# Functions
+
+##
+# Parses one line +ln+ in SAM format and outputs filtered lines to +ofh+
+# Filters by minimum +identity+ and +target+ sequences, and prints
+# the headers if +header+
+def parse_sam_line(ln, identity, target, header, ofh)
+  if ln =~ /^@/ || ln =~ /^\s*$/
+    ofh.puts ln if header
+    return
+  end
+
+  # No match
+  row = ln.chomp.split("\t")
+  return if row[2] == '*'
+
+  # Filter by target
+  return if !target.nil? && !target.include?(row[2])
+
+  # Exclude unless concordant or unaligned
+  length = row[9].size
+  row.shift(11) # Discard non-flag columns
+  flags = Hash[row.map { |i| i.sub(/:.:/, ':').split(':', 2) }]
+  return if flags['YT'] && !%w[CP UU].include?(flags['YT'])
+
+  # Filter by identity
+  unless flags['MD']
+    raise Enveomics::ParseError.new(
+      "SAM line missing MD flag:\n#{ln}\nFlags: #{flags}"
+    )
+  end
+  mismatches = flags['MD'].scan(/[^\d]/).count
+  id = 100.0 * (length - mismatches) / length
+  ofh.puts ln if id >= identity
+end
+
+# Reading targets
+if o[:g]
+  say 'Loading target sequences to filter'
+  reader = reader(o[:g])
+  target =
+    case o[:g_format]
+    when :fasta
+      reader.each.map { |ln| $1 if ln =~ /^>(\S+)/ }.compact
+    when :list
+      reader.each.map(&:chomp)
+    else
+      raise Enveomics::OptionError.new(
+        "Unsupported target sequences format: #{o[:g_format]}"
+      )
+    end
+  reader.close
+else
+  target = nil
+end
+
+# Reading and filtering mapping
+say 'Reading mapping file'
+ofh = writer(o[:o])
+case o[:m_format]
+when :sam
+  reader = reader(o[:m])
+  reader.each { |ln| parse_sam_line(ln, o[:identity], target, o[:header], ofh) }
+  reader.close
+when :bam
+  cmd = ['samtools', 'view', o[:m], '-@', o[:threads]]
+  cmd << '-h' if o[:header]
+  IO.popen(cmd.shelljoin) do |fh|
+    fh.each { |ln| parse_sam_line(ln, o[:identity], target, o[:header], ofh) }
+  end
+else
+  raise Enveomics::OptionError.new(
+    "Unsupported mapping format: #{o[:m_format]}"
+  )
+end
+ofh.close
+

data/utils/enveomics/enveomics.R/DESCRIPTION

@@ -1,5 +1,5 @@
 Package: enveomics.R
-Version: 1.7.1
+Version: 1.8.0
 Authors@R: c(person("Luis M.","Rodriguez-R",role=c("aut","cre"),
 	   email="lmrodriguezr@gmail.com"))
 Title: Various Utilities for Microbial Genomics and Metagenomics
@@ -28,4 +28,4 @@ Suggests:
 License: Artistic-2.0
 LazyData: yes
 Encoding: UTF-8
-RoxygenNote: 6.1.1
+RoxygenNote: 7.0.2

data/utils/enveomics/enveomics.R/NAMESPACE

@@ -36,4 +36,4 @@ export(
   enve.recplot2.compareIdentities,
   enve.recplot2.coordinates, enve.recplot2.seqdepth, enve.recplot2.ANIr,
   enve.growthcurve, plot.enve.GrowthCurve, summary.enve.GrowthCurve,
-  enve.col2alpha, enve.truncate)
+  enve.col2alpha, enve.truncate, enve.selvector, enve.prefscore)

data/utils/enveomics/enveomics.R/R/prefscore.R

@@ -0,0 +1,79 @@
+#' Enveomics: Pref Score
+#'
+#' Estimate preference score of species based on occupancy in biased sample sets
+#'
+#' @param x
+#' Occupancy matrix (logical or numeric binary) with species as rows and samples
+#' as columns
+#' @param set
+#' Vector indicating samples in the test set. It can be any selection vector:
+#' boolean (same length as the number of columns in \code{x}), or numeric or
+#' character vector with indexes of the \code{x} columns.
+#' @param ignore
+#' Vector indicating species to ignore. It can be any selection vector with
+#' respect to the rows in \code{x} (see \code{set}).
+#' @param signif.thr Absolute value of the significance threshold
+#' @param plot Indicates if a plot should be generated
+#' @param col.above Color for points significantly above zero
+#' @param col.equal Color for points not significantly different from zero
+#' @param col.below Color for points significantly below zero
+#' @param ... Any additional parameters supported by \code{plot}
+#'
+#' @return Returns a named vector of preference scores.
+#'
+#' @author Luis M. Rodriguez-R [aut, cre]
+#'
+#' @export
+
+enve.prefscore <- function
+(
+  x,
+  set,
+  ignore = NULL,
+  signif.thr,
+  plot = TRUE,
+  col.above = rgb(148, 17, 0, maxColorValue = 255),
+  col.equal = rgb(189, 189, 189, maxColorValue = 255),
+  col.below = rgb(47, 84, 150, maxColorValue = 255),
+  ...
+) {
+  # Normalize classes and filter universe
+  x <- !!as.matrix(x)
+  if(is.null(colnames(x))) colnames(x) <- 1:ncol(x)
+  if(is.null(rownames(x))) rownames(x) <- 1:nrow(x)
+  set <- enve.selvector(set, colnames(x))
+  universe <- !enve.selvector(ignore, rownames(x))
+  x.u <- x[universe, ]
+  if(missing(signif.thr)) signif.thr <- 1 + 100 / length(universe)
+
+  # Base (null) probabilities
+  p_a <- (rowSums(x.u) + 1) / (ncol(x.u) + 2)
+  p_b <- (colSums(x.u) + 1) / (nrow(x.u) + 2)
+  p_p <- p_a %*% t(p_b)
+
+  # Set preference score
+  expected <- (rowSums(p_p[, set]) - rowSums(p_p[, !set])) / sum(p_p)
+  observed <- (rowSums(x.u[, set]) - rowSums(x.u[, !set])) / sum(x.u)
+  y <- observed / abs(expected)
+  names(y) <- rownames(x)[universe]
+  y.code <- cut(y, c(-Inf, -signif.thr, signif.thr, Inf), 1:3)
+
+  # Plot
+  if(plot) {
+    idx <- (1:nrow(x))[universe]
+    opts.def <- list(x = idx, y = y, ylim = c(-1, 1) * max(abs(y)),
+      xlab = 'Species', ylab = 'Preference score', xlim = c(0, nrow(x)+1),
+      col = c(col.above, col.equal, col.below)[y.code],
+      las = 1, xaxs = 'i', pch = 15)
+    opts <- list(...)
+    for(i in names(opts.def)) if(is.null(opts[[i]])) opts[[i]] <- opts.def[[i]]
+    do.call('plot', opts)
+    abline(h = 0, lty = 1, col = rgb(0, 0, 0, 1/4))
+    abline(h = c(-1, 1) * signif.thr, lty = 2, col = rgb(0, 0, 0, 1/4))
+  }
+
+  # Print and return
+  print(table(c(c('<', '=', '>')[y.code], rep('Tot', length(y.code)))))
+  cat('---------\n')
+  return(y)
+}

data/utils/enveomics/enveomics.R/R/utils.R

@@ -48,3 +48,33 @@ enve.truncate <- function
   y <- sort(x)[ -c(seq(1, n), seq(length(x)+1-n, length(x))) ]
   return(FUN(y))
 }
+
+#' Enveomics: Selection vector
+#' 
+#' Normalizes a selection vector \code{sel} to a logical vector with indexes
+#' from \code{dim.names}.
+#'
+#' @param sel A vector of numbers, characters, or booleans.
+#' @param dim.names A vector of names from which to select.
+#' 
+#' @return Returns a logical vector with the same length as \code{dim.name}.
+#' 
+#' @author Luis M. Rodriguez-R [aut, cre]
+#'
+#' @export
+
+enve.selvector <- function(sel, dim.names) {
+  if(is.logical(sel)) {
+    if(length(sel) != length(dim.names))
+      stop('sel is logical but differs in length from dim.names')
+    sel
+  } else if(is.numeric(sel)) {
+    if(max(sel) > length(dim.names))
+      stop('sel includes numeric index beyond the length of dim.names')
+    1:length(dim.names) %in% sel
+  } else {
+    if(any(!sel %in% dim.names))
+      stop('sel includes character index missing from dim.names')
+    dim.names %in% sel
+  }
+}

data/utils/enveomics/enveomics.R/README.md

@@ -52,6 +52,7 @@ For additional information on recruitment plots, see the
 [Recruitment plots working document](https://github.com/lmrodriguezr/enveomics/blob/master/Docs/recplot2.md).
 
 ## Changelog
+* 1.8.0: New functions `enve.selvector` and `enve.prefscore`.
 * 1.7.1: Improved efficiency of `enve.df2dist` about five-fold.
 * 1.7.0: Uniformized output for `enve.recplot2.extractWindows` and
   `enve.recplot2.coordinates` to ease automation. Thanks to Tomeu Viver and

data/utils/enveomics/enveomics.R/man/cash-enve.GrowthCurve-method.Rd

@@ -1,6 +1,5 @@
 % Generated by roxygen2: do not edit by hand
 % Please edit documentation in R/growthcurve.R
-\docType{methods}
 \name{$,enve.GrowthCurve-method}
 \alias{$,enve.GrowthCurve-method}
 \title{Attribute accessor}

data/utils/enveomics/enveomics.R/man/cash-enve.RecPlot2-method.Rd

@@ -1,6 +1,5 @@
 % Generated by roxygen2: do not edit by hand
 % Please edit documentation in R/recplot2.R
-\docType{methods}
 \name{$,enve.RecPlot2-method}
 \alias{$,enve.RecPlot2-method}
 \title{Attribute accessor}

data/utils/enveomics/enveomics.R/man/cash-enve.RecPlot2.Peak-method.Rd

@@ -1,6 +1,5 @@
 % Generated by roxygen2: do not edit by hand
 % Please edit documentation in R/recplot2.R
-\docType{methods}
 \name{$,enve.RecPlot2.Peak-method}
 \alias{$,enve.RecPlot2.Peak-method}
 \title{Attribute accessor}

data/utils/enveomics/enveomics.R/man/enve.__tribs.Rd

@@ -4,8 +4,16 @@
 \alias{enve.__tribs}
 \title{Enveomics: TRIBS - Internal Ancillary Function}
 \usage{
-enve.__tribs(rep, frx, selection, dimensions, dots, dist.method,
-  summary.fx, dist)
+enve.__tribs(
+  rep,
+  frx,
+  selection,
+  dimensions,
+  dots,
+  dist.method,
+  summary.fx,
+  dist
+)
 }
 \arguments{
 \item{rep}{Replicates}

data/utils/enveomics/enveomics.R/man/enve.barplot.Rd

@@ -4,10 +4,22 @@
 \alias{enve.barplot}
 \title{Enveomics: Barplot}
 \usage{
-enve.barplot(x, sizes, top = 25, colors.per.group = 9,
-  bars.width = 4, legend.ncol = 1, other.col = "#000000",
-  add.trend = FALSE, organic.trend = FALSE, sort.by = median,
-  min.report = 101, order = NULL, col, ...)
+enve.barplot(
+  x,
+  sizes,
+  top = 25,
+  colors.per.group = 9,
+  bars.width = 4,
+  legend.ncol = 1,
+  other.col = "#000000",
+  add.trend = FALSE,
+  organic.trend = FALSE,
+  sort.by = median,
+  min.report = 101,
+  order = NULL,
+  col,
+  ...
+)
 }
 \arguments{
 \item{x}{Can be either the input data or the path to the file containing 

data/utils/enveomics/enveomics.R/man/enve.cliopts.Rd

@@ -4,9 +4,19 @@
 \alias{enve.cliopts}
 \title{Enveomics: Cliopts}
 \usage{
-enve.cliopts(fx, rd_file, positional_arguments, usage, mandatory = c(),
-  vectorize = c(), ignore = c(), number = c(), defaults = list(),
-  o_desc = list(), p_desc = "")
+enve.cliopts(
+  fx,
+  rd_file,
+  positional_arguments,
+  usage,
+  mandatory = c(),
+  vectorize = c(),
+  ignore = c(),
+  number = c(),
+  defaults = list(),
+  o_desc = list(),
+  p_desc = ""
+)
 }
 \arguments{
 \item{fx}{Function for which the interface should be generated.}

data/utils/enveomics/enveomics.R/man/enve.df2dist.Rd

@@ -4,8 +4,14 @@
 \alias{enve.df2dist}
 \title{Enveomics: Data Frame to Dist}
 \usage{
-enve.df2dist(x, obj1.index = 1, obj2.index = 2, dist.index = 3,
-  default.d = NA, max.sim = 0)
+enve.df2dist(
+  x,
+  obj1.index = 1,
+  obj2.index = 2,
+  dist.index = 3,
+  default.d = NA,
+  max.sim = 0
+)
 }
 \arguments{
 \item{x}{A dataframe (or coercible object) with at least three columns:

data/utils/enveomics/enveomics.R/man/enve.df2dist.group.Rd

@@ -4,8 +4,14 @@
 \alias{enve.df2dist.group}
 \title{Enveomics: Data Frame to Dist (Group)}
 \usage{
-enve.df2dist.group(x, obj1.index = 1, obj2.index = 2, dist.index = 3,
-  summary = median, empty.rm = TRUE)
+enve.df2dist.group(
+  x,
+  obj1.index = 1,
+  obj2.index = 2,
+  dist.index = 3,
+  summary = median,
+  empty.rm = TRUE
+)
 }
 \arguments{
 \item{x}{A dataframe (or coercible object) with at least three columns:

data/utils/enveomics/enveomics.R/man/enve.df2dist.list.Rd

@@ -4,8 +4,15 @@
 \alias{enve.df2dist.list}
 \title{Enveomics: Data Frame to Dist (List)}
 \usage{
-enve.df2dist.list(x, groups, obj1.index = 1, obj2.index = 2,
-  dist.index = 3, empty.rm = TRUE, ...)
+enve.df2dist.list(
+  x,
+  groups,
+  obj1.index = 1,
+  obj2.index = 2,
+  dist.index = 3,
+  empty.rm = TRUE,
+  ...
+)
 }
 \arguments{
 \item{x}{A dataframe (or coercible object) with at least three columns:

data/utils/enveomics/enveomics.R/man/enve.growthcurve.Rd

@@ -4,11 +4,19 @@
 \alias{enve.growthcurve}
 \title{Enveomics: Growth Curve}
 \usage{
-enve.growthcurve(x, times = 1:nrow(x), triplicates = FALSE, design,
-  new.times = seq(min(times), max(times), length.out = length(times) *
-  10), level = 0.95, interval = c("confidence", "prediction"),
-  plot = TRUE, FUN = function(t, K, r, P0) K * P0 * exp(r * t)/(K + P0
-  * (exp(r * t) - 1)), nls.opt = list(), ...)
+enve.growthcurve(
+  x,
+  times = 1:nrow(x),
+  triplicates = FALSE,
+  design,
+  new.times = seq(min(times), max(times), length.out = length(times) * 10),
+  level = 0.95,
+  interval = c("confidence", "prediction"),
+  plot = TRUE,
+  FUN = function(t, K, r, P0) K * P0 * exp(r * t)/(K + P0 * (exp(r * t) - 1)),
+  nls.opt = list(),
+  ...
+)
 }
 \arguments{
 \item{x}{Data frame (or coercible) containing the observed growth data