bio-maf 0.1.0

data/lib/bio/maf/struct.rb

@@ -0,0 +1,63 @@
+## NOTE: this is probably not the best place for this, ultimately.
+## If it works, think about moving it.
+
+module Bio
+
+  module MAF
+
+    class Struct
+      def initialize(spec)
+        @members = []
+        @by_name = {}
+        offset = 0
+        spec.each do |m_spec|
+          m = Member.new(offset, *m_spec)
+          @members << m
+          @by_name[m.name] = m
+          offset += m.size
+        end
+      end
+
+      def fmt
+        @members.collect { |m| m.fmt }.join('')
+      end
+
+      def extractor_fmt(*names)
+        extract = names.collect { |name| @by_name.fetch(name) }
+        extract.sort_by! { |m| m.offset }
+        fmt = ''
+        pos = 0
+        extract.each do |member|
+          if member.offset != pos
+            fmt << "@#{member.offset}"
+            pos = member.offset
+          end
+          fmt << member.fmt
+          pos += member.size
+        end
+        return fmt
+      end
+    end
+
+    TYPE_PROPS = {
+      :uint8  => { :size => 1, :fmt => 'C'  },
+      :uint16 => { :size => 2, :fmt => 'S>' },
+      :uint32 => { :size => 4, :fmt => 'L>' },
+      :uint64 => { :size => 8, :fmt => 'Q>' }
+    }
+
+    class Member
+      attr_reader :offset, :name, :type, :size, :fmt
+      def initialize(offset, name, type)
+        @offset = offset
+        @name = name
+        @type = type
+        props = TYPE_PROPS.fetch(type)
+        @size = props.fetch(:size)
+        @fmt = props.fetch(:fmt)
+      end
+    end
+
+ end
+  
+end

data/lib/bio/maf/writer.rb

@@ -0,0 +1,63 @@
+module Bio::MAF
+
+  class Writer
+    attr_reader :f, :path
+
+    def initialize(fspec)
+      if fspec.respond_to? :write
+        @f = fspec
+        if fspec.respond_to? :path
+          @path = fspec.path
+        end
+      else
+        @path = fspec
+        @f = File.open(fspec, 'w')
+      end
+    end
+
+    def flatten_vars(vars)
+      vars.to_a.collect {|k, v| "#{k}=#{v}"}.join(" ")
+    end
+
+    def write_header(header)
+      f.puts "##maf #{flatten_vars(header.vars)}"
+      f.puts "##{header.alignment_params}" if header.alignment_params
+    end
+
+    def write_blocks(blocks)
+      blocks.each do |block|
+        write_block(block)
+      end
+      f.flush
+    end
+
+    def write_block(block)
+      lines = ["a #{flatten_vars(block.vars)}"]
+      block.sequences.each do |seq| 
+        write_seq(seq, lines)
+      end
+      lines << " "
+      f.puts lines.join("\n")
+    end
+
+    def write_seq(s, lines)
+      lines << sprintf("%s %-20s %12d %2d %s %9d %s",
+                       s.empty? ? "e" : "s",
+                       s.source,
+                       s.start,
+                       s.size,
+                       s.strand,
+                       s.src_size,
+                       s.empty? ? s.status : s.text)
+      if s.quality
+        lines << sprintf("q %-20s                           %s",
+                         s.source, s.quality)
+      end
+      if s.i_data
+        lines << sprintf("i %-20s %s %s %s %s",
+                         s.source, *s.i_data)
+      end
+    end
+  end
+  
+end

data/lib/bio/maf.rb

@@ -0,0 +1,4 @@
+require 'bio/maf/struct'
+require 'bio/maf/index'
+require 'bio/maf/parser'
+require 'bio/maf/writer'

data/lib/bio/ucsc/genomic-interval-bin.rb

@@ -0,0 +1,13 @@
+require "bio-genomic-interval"
+
+module Bio
+  class GenomicInterval
+    def bin
+      Bio::Ucsc::UcscBin.bin(self.zero_start, self.zero_end)
+    end
+
+    def bin_all
+      Bio::Ucsc::UcscBin.bin_all(self.zero_start, self.zero_end)
+    end
+  end
+end

data/lib/bio/ucsc/ucsc_bin.rb

@@ -0,0 +1,117 @@
+# = UCSCBin
+# Author::    MISHIMA, Hiroyuki
+# Copyright:: MISHIMA, Hiroyuki, 2010-2011
+# License::   The Ruby licence (Ryby's / GPLv2 dual)
+#
+# Original program in C by Jim Kent, 2002
+# See also http://genomewiki.ucsc.edu/index.php/Bin_indexing_system;
+# a paper Kent, et. al. Genome Research 2002.12:996-1006;
+# and src/lib/binRange.c in the kent source tree.
+#
+# Bio::Ucsc::UcscBin - 
+# 1) convert between 0-based half-open interval and
+#    1-based full-close intervals.
+# 2) Calculate Bin number from genomic physical position
+# according to UCSC's Bin Indexing System. 
+#
+module Bio
+  module Ucsc
+    class UcscBin
+      # Version = "0.1.0" # 20100714
+      # Version = "0.2.0" # 20101028
+      # Version = "0.2.1" # 20110408
+      Version = "0.2.2" # 20110418 the licence is changed 
+                        # embeded in BioRubyUcscApi
+                        # handle the case, start==end in [start, end) 
+      
+      BINRANGE_MAXEND_512M       = (512*1024*1024)
+      BIN_OFFSETS_EXTENDED       = [4096+512+64+8+1, 512+64+8+1, 64+8+1, 8+1, 1, 0]
+ #     BIN_OFFSETS_EXTENDED = [4096+512+64+8+1, 512+64+8+1, 64+8+1, 8+1, 1] (to omit BIN=0)
+      BIN_OFFSETS                = [512+64+8+1, 64+8+1, 8+1, 1, 0]
+ #     BIN_OFFSETS = [512+64+8+1, 64+8+1, 8+1, 1] (to omit BIN=0)
+      BIN_OFFSET_OLD_TO_EXTENDED = 4681
+      # How much to shift to get to finest bin.
+      BIN_FIRST_SHIFT            = 17 
+      # How much to shift to get to next larger bin.
+      BIN_NEXT_SHIFT             = 3 
+
+      # Return a Integer of a BIN which is the smallest/finest bin 
+      # containing whole the interval/range.
+      # 
+      # Extended bin index for positions >= 512M is not supported yet
+      # Do you need it? Please email me.
+      def self.bin_from_range(bin_start, bin_end)
+        if bin_end <= BINRANGE_MAXEND_512M
+          bin_from_range_standard(bin_start, bin_end)
+        else
+          bin_from_range_extended(bin_start, bin_end)
+        end
+      end
+      
+      class << self;  alias bin bin_from_range; end
+
+      # Return an Array of BINs which are all bins containing whole the
+      # interval/range. Thus, it always contains "0" indicating a bin
+      # containing whole of a chromosome.
+      # 
+      # extended bin index for positions >= 512M is not supported yet
+      # Do you need it? Please email me.
+      #
+      def self.bin_all(p_start, p_end)
+        if p_end <= BINRANGE_MAXEND_512M
+          bin_all_standard(p_start, p_end)
+        else
+          bin_all_extended(p_start, p_end)
+        end
+      end
+
+      private
+
+      def self.bin_from_range_standard(bin_start, bin_end)
+        # Given start,end in chromosome coordinates assign it
+        # a bin.   There's a bin for each 128k segment, for each
+        # 1M segment, for each 8M segment, for each 64M segment,
+        # and for each chromosome (which is assumed to be less than
+        # 512M.)  A range goes into the smallest bin it will fit in.
+
+        bin_start >>= BIN_FIRST_SHIFT
+        bin_end -= 1
+        bin_end >>= BIN_FIRST_SHIFT
+
+        BIN_OFFSETS.each do |offset|
+          return offset + bin_start if bin_start == bin_end
+          bin_start >>= BIN_NEXT_SHIFT
+          bin_end   >>= BIN_NEXT_SHIFT
+        end
+        raise RangeError, \
+        "start #{bin_start}, end #{bin_end} out of range in findBin (max is 512M)"
+      end
+
+      def self.bin_from_range_extended(bin_start, bin_end)
+        raise NotImplementedError, "Extended bins are not supported yet"
+      end
+
+      def self.bin_all_standard(bin_start, bin_end)
+        bin_start_orig = bin_start
+        bin_end_orig   = bin_end 
+        results = Array.new
+        
+        bin_start >>= BIN_FIRST_SHIFT
+        bin_end -= 1
+        bin_end >>= BIN_FIRST_SHIFT
+
+        BIN_OFFSETS.each do |offset|
+          results.concat(((offset + bin_start)..(offset + bin_end)).to_a)
+          bin_start >>= BIN_NEXT_SHIFT
+          bin_end   >>= BIN_NEXT_SHIFT
+        end
+        return results
+      end
+      
+      def self.bin_all_extended(bin_start, bin_end)
+        raise NotImplementedError, "Extended bins are not supported yet"
+      end
+
+    end # class UcscBin
+  end # module Ucsc
+end # module Bio 

data/lib/bio/ucsc.rb

@@ -0,0 +1,2 @@
+require 'bio/ucsc/ucsc_bin'
+require 'bio/ucsc/genomic-interval-bin'

data/lib/bio-maf/maf.rb

@@ -0,0 +1,3 @@
+module BioMaf
+
+end

data/lib/bio-maf.rb

@@ -0,0 +1,12 @@
+# Please require your code below, respecting the naming conventions in the
+# bioruby directory tree.
+#
+# For example, say you have a plugin named bio-plugin, the only uncommented
+# line in this file would be 
+#
+#   require 'bio/bio-plugin/plugin'
+#
+# In this file only require other files. Avoid other source code.
+
+require 'bio/ucsc'
+require 'bio/maf'

data/man/.gitignore

@@ -0,0 +1 @@
+*.html

data/man/maf_index.1

@@ -0,0 +1,105 @@
+.\" generated with Ronn/v0.7.3
+.\" http://github.com/rtomayko/ronn/tree/0.7.3
+.
+.TH "MAF_INDEX" "1" "June 2012" "Clayton Wheeler" "BioRuby Manual"
+.
+.SH "NAME"
+\fBmaf_index\fR \- build and examine MAF indexes
+.
+.SH "SYNOPSIS"
+\fBmaf_index\fR [\-t] \fImaf\fR \fIindex\fR
+.
+.br
+\fBmaf_index\fR \fB\-d\fR|\fB\-\-dump\fR \fIindex\fR
+.
+.SH "DESCRIPTION"
+\fBmaf_index\fR is part of the bioruby\-maf library and creates Kyoto Cabinet indexes for Multiple Alignment Format (MAF) files\. These indexes enable other MAF tools to selectively extract alignment blocks of interest\.
+.
+.P
+In its default mode, \fBmaf_index\fR parses the \fImaf\fR file given as an argument and creates an index in \fIindex\fR\.
+.
+.P
+The index data is stored in binary form, so with the \fB\-\-dump\fR argument, \fBmaf_index\fR can dump out the index data in human\-readable form for debugging\.
+.
+.SH "FILES"
+The \fImaf\fR input file must be a valid MAF file of any length\.
+.
+.P
+The index created is a Kyoto Cabinet TreeDB (B+ tree) database; \fIindex\fR must have a \fB\.kct\fR extension\.
+.
+.SH "OPTIONS"
+TODO
+.
+.TP
+\fB\-d\fR, \fB\-\-dump\fR
+Instead of creating an index, dump out the given \fIindex\fR in human\-readable form\. Index records will appear like:
+.
+.IP "" 4
+.
+.nf
+
+0 [bin 1195] 80082334:80082368
+  offset 16, length 1087
+  text size: 54
+  sequences in block: 10
+  species vector: 00000000000003ff
+.
+.fi
+.
+.IP "" 0
+
+.
+.TP
+\fB\-t\fR, \fB\-\-threaded\fR
+Use a separate reader thread to do I/O in parallel with parsing\. Only useful on JRuby\.
+.
+.TP
+\fB\-\-time\fR
+Print elapsed time for index creation\. Mainly useful for measuring performance with different Ruby implementations, I/O subsystems, etc\.
+.
+.SH "EXAMPLES"
+Build an index on a MAF file:
+.
+.IP "" 4
+.
+.nf
+
+$ maf_index chr22\.maf chr22\.kct
+.
+.fi
+.
+.IP "" 0
+.
+.P
+Dump out an index:
+.
+.IP "" 4
+.
+.nf
+
+$ maf_index \-d chr22\.kct > /tmp/chr22\.dump
+.
+.fi
+.
+.IP "" 0
+.
+.SH "ENVIRONMENT"
+\fBmaf_index\fR is a Ruby program and relies on ordinary Ruby environment variables\.
+.
+.SH "BUGS"
+\fBmaf_index\fR does not currently allow Kyoto Cabinet database parameters to be set\.
+.
+.SH "COPYRIGHT"
+\fBmaf_index\fR is copyright (C) 2012 Clayton Wheeler\.
+.
+.SH "SEE ALSO"
+ruby(1), kctreemgr(1)
+.
+.IP "\(bu" 4
+\fIhttps://github\.com/csw/bioruby\-maf/\fR
+.
+.IP "\(bu" 4
+\fIhttp://fallabs\.com/kyotocabinet/\fR
+.
+.IP "" 0
+

data/man/maf_index.1.markdown

@@ -0,0 +1,97 @@
+maf_index(1) -- build and examine MAF indexes
+=============================================
+
+## SYNOPSIS
+
+`maf_index` [-t] <var>maf</var> <var>index</var><br>
+`maf_index` `-d`|`--dump` <var>index</var>
+
+## DESCRIPTION
+
+**maf_index** is part of the bioruby-maf library and creates
+Kyoto Cabinet indexes for Multiple Alignment Format (MAF)
+files. These indexes enable other MAF tools to selectively extract
+alignment blocks of interest.
+
+In its default mode, `maf_index` parses the <var>maf</var> file given as an
+argument and creates an index in <var>index</var>. 
+
+The index data is stored in binary form, so with the `--dump`
+argument, `maf_index` can dump out the index data in human-readable
+form for debugging.
+
+## FILES
+
+The <var>maf</var> input file must be a valid MAF file of any length.
+
+The index created is a Kyoto Cabinet TreeDB (B+ tree) database;
+<var>index</var> must have a `.kct` extension.
+
+## OPTIONS
+
+TODO
+
+ * `-d`, `--dump`:
+   Instead of creating an index, dump out the given <var>index</var> in
+   human-readable form. Index records will appear like:
+   
+       0 [bin 1195] 80082334:80082368
+         offset 16, length 1087
+         text size: 54
+         sequences in block: 10
+         species vector: 00000000000003ff
+
+ * `-t`, `--threaded`:
+   Use a separate reader thread to do I/O in parallel with
+   parsing. Only useful on JRuby.
+
+ * `--time`:
+   Print elapsed time for index creation. Mainly useful for measuring
+   performance with different Ruby implementations, I/O subsystems,
+   etc.
+   
+## EXAMPLES
+
+Build an index on a MAF file:
+
+    $ maf_index chr22.maf chr22.kct
+    
+Dump out an index:
+
+    $ maf_index -d chr22.kct > /tmp/chr22.dump
+
+## ENVIRONMENT
+
+`maf_index` is a Ruby program and relies on ordinary Ruby environment
+variables.
+
+## BUGS
+
+`maf_index` does not currently allow Kyoto Cabinet database parameters
+to be set.
+
+## COPYRIGHT
+
+`maf_index` is copyright (C) 2012 Clayton Wheeler.
+
+## SEE ALSO
+
+ruby(1), kctreemgr(1)
+
+ * <https://github.com/csw/bioruby-maf/>
+ * <http://fallabs.com/kyotocabinet/>
+
+
+
+[SYNOPSIS]: #SYNOPSIS "SYNOPSIS"
+[DESCRIPTION]: #DESCRIPTION "DESCRIPTION"
+[FILES]: #FILES "FILES"
+[OPTIONS]: #OPTIONS "OPTIONS"
+[EXAMPLES]: #EXAMPLES "EXAMPLES"
+[ENVIRONMENT]: #ENVIRONMENT "ENVIRONMENT"
+[BUGS]: #BUGS "BUGS"
+[COPYRIGHT]: #COPYRIGHT "COPYRIGHT"
+[SEE ALSO]: #SEE-ALSO "SEE ALSO"
+
+
+[maf_index(1)]: maf_index.1.html

data/man/maf_index.1.ronn

@@ -0,0 +1,83 @@
+maf_index(1) -- build and examine MAF indexes
+=============================================
+
+## SYNOPSIS
+
+`maf_index` [-t] <maf> <index><br>
+`maf_index` `-d`|`--dump` <index>
+
+## DESCRIPTION
+
+**maf_index** is part of the bioruby-maf library and creates
+Kyoto Cabinet indexes for Multiple Alignment Format (MAF)
+files. These indexes enable other MAF tools to selectively extract
+alignment blocks of interest.
+
+In its default mode, `maf_index` parses the <maf> file given as an
+argument and creates an index in <index>. 
+
+The index data is stored in binary form, so with the `--dump`
+argument, `maf_index` can dump out the index data in human-readable
+form for debugging.
+
+## FILES
+
+The <maf> input file must be a valid MAF file of any length.
+
+The index created is a Kyoto Cabinet TreeDB (B+ tree) database;
+<index> must have a `.kct` extension.
+
+## OPTIONS
+
+TODO
+
+ * `-d`, `--dump`:
+   Instead of creating an index, dump out the given <index> in
+   human-readable form. Index records will appear like:
+   
+       0 [bin 1195] 80082334:80082368
+         offset 16, length 1087
+         text size: 54
+         sequences in block: 10
+         species vector: 00000000000003ff
+
+ * `-t`, `--threaded`:
+   Use a separate reader thread to do I/O in parallel with
+   parsing. Only useful on JRuby.
+
+ * `--time`:
+   Print elapsed time for index creation. Mainly useful for measuring
+   performance with different Ruby implementations, I/O subsystems,
+   etc.
+   
+## EXAMPLES
+
+Build an index on a MAF file:
+
+    $ maf_index chr22.maf chr22.kct
+    
+Dump out an index:
+
+    $ maf_index -d chr22.kct > /tmp/chr22.dump
+
+## ENVIRONMENT
+
+`maf_index` is a Ruby program and relies on ordinary Ruby environment
+variables.
+
+## BUGS
+
+`maf_index` does not currently allow Kyoto Cabinet database parameters
+to be set.
+
+## COPYRIGHT
+
+`maf_index` is copyright (C) 2012 Clayton Wheeler.
+
+## SEE ALSO
+
+ruby(1), kctreemgr(1)
+
+ * <https://github.com/csw/bioruby-maf/>
+ * <http://fallabs.com/kyotocabinet/>
+

data/man/maf_to_fasta.1

@@ -0,0 +1,53 @@
+.\" generated with Ronn/v0.7.3
+.\" http://github.com/rtomayko/ronn/tree/0.7.3
+.
+.TH "MAF_TO_FASTA" "1" "June 2012" "Clayton Wheeler" "BioRuby Manual"
+.
+.SH "NAME"
+\fBmaf_to_fasta\fR \- convert MAF file to FASTA
+.
+.SH "SYNOPSIS"
+\fBmaf_to_fasta\fR [\fIoptions\fR\.\.\.] \fImaf\fR \fIfasta\fR
+.
+.SH "DESCRIPTION"
+\fBmaf_to_fasta\fR, part of the bioruby\-maf library, converts Multiple Alignment Format (MAF) files to FASTA format\. It does not attempt to combine alignment blocks, but simply writes out each sequence in order\.
+.
+.SH "FILES"
+The \fImaf\fR input file must be a valid MAF file of any length\.
+.
+.P
+The \fIfasta\fR output file will be written in FASTA format\.
+.
+.SH "OPTIONS"
+The options are only useful for performance measurement\.
+.
+.SH "EXAMPLES"
+Convert a MAF file to FASTA:
+.
+.IP "" 4
+.
+.nf
+
+$ maf_to_fasta chrY\.maf chrY\.fa
+.
+.fi
+.
+.IP "" 0
+.
+.SH "ENVIRONMENT"
+\fBmaf_to_fasta\fR is a Ruby program and relies on ordinary Ruby environment variables\.
+.
+.SH "BUGS"
+\fBmaf_to_fasta\fR should provide flexibility in selecting the alignment blocks and sequences to convert, and perhaps allow alignment blocks to be spliced together\.
+.
+.SH "COPYRIGHT"
+\fBmaf_to_fasta\fR is copyright (C) 2012 Clayton Wheeler\.
+.
+.SH "SEE ALSO"
+ruby(1)
+.
+.IP "\(bu" 4
+\fIhttps://github\.com/csw/bioruby\-maf/\fR
+.
+.IP "" 0
+

data/man/maf_to_fasta.1.ronn

@@ -0,0 +1,51 @@
+maf_to_fasta(1) -- convert MAF file to FASTA
+============================================
+
+## SYNOPSIS
+
+`maf_to_fasta` [<options>...] <maf> <fasta>
+
+## DESCRIPTION
+
+**maf_to_fasta**, part of the bioruby-maf library, converts Multiple
+Alignment Format (MAF) files to FASTA format. It does not attempt to
+combine alignment blocks, but simply writes out each sequence in
+order.
+
+## FILES
+
+The <maf> input file must be a valid MAF file of any length.
+
+The <fasta> output file will be written in FASTA format.
+
+## OPTIONS
+
+The options are only useful for performance measurement.
+
+## EXAMPLES
+
+Convert a MAF file to FASTA:
+
+    $ maf_to_fasta chrY.maf chrY.fa
+    
+## ENVIRONMENT
+
+`maf_to_fasta` is a Ruby program and relies on ordinary Ruby
+environment variables.
+
+## BUGS
+
+`maf_to_fasta` should provide flexibility in selecting the alignment
+blocks and sequences to convert, and perhaps allow alignment blocks to
+be spliced together. 
+
+## COPYRIGHT
+
+`maf_to_fasta` is copyright (C) 2012 Clayton Wheeler.
+
+## SEE ALSO
+
+ruby(1)
+
+ * <https://github.com/csw/bioruby-maf/>
+