1 # Copyright (C) 2007-2012 Martin A. Hansen.
3 # This program is free software; you can redistribute it and/or
4 # modify it under the terms of the GNU General Public License
5 # as published by the Free Software Foundation; either version 2
6 # of the License, or (at your option) any later version.
8 # This program is distributed in the hope that it will be useful,
9 # but WITHOUT ANY WARRANTY; without even the implied warranty of
10 # MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the
11 # GNU General Public License for more details.
13 # You should have received a copy of the GNU General Public License
14 # along with this program; if not, write to the Free Software
15 # Foundation, Inc., 51 Franklin Street, Fifth Floor, Boston, MA 02110-1301, USA.
17 # http://www.gnu.org/copyleft/gpl.html
19 # >>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>><<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<
21 # This software is part of the Biopieces framework (www.biopieces.org).
23 # >>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>><<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<
25 require 'maasha/align/match'
26 require 'maasha/math_aux'
28 FACTOR_SCORE_LENGTH = 1.0
29 FACTOR_SCORE_DIAG = -1.42
31 # Module with stuff to create a pairwise aligment.
33 # Class for creating a pairwise alignment.
35 # Class method to create a pairwise alignment of two given Seq objects.
36 def self.align(q_entry, s_entry)
37 self.new(q_entry, s_entry)
40 # Method to inialize a pairwise alignment given two Seq objects.
41 def initialize(q_entry, s_entry)
46 @q_entry.seq.downcase!
47 @s_entry.seq.downcase!
49 space = Space.new( 0, 0, @q_entry.length - 1, @s_entry.length - 1)
52 align_recurse(@q_entry.seq, @s_entry.seq, space, kmer)
59 # Method that creates an alignment by chaining matches, which are
60 # subsequences shared between two sequences. This recursive method
61 # functions by considering only matches within a given search space. If no
62 # matches are given these will be located and matches will be included
63 # depending on a calculated score. New search spaces spanning the spaces
64 # between the best scoring matches and the search space boundaries will be
65 # cast and recursed into.
66 def align_recurse(q_seq, s_seq, space, kmer, matches = [])
67 matches = matches_select_by_space(matches, space)
69 while (matches.size == 0 and kmer > 0)
70 matches = Matches.find(q_seq, s_seq, space.q_min, space.s_min, space.q_max, space.s_max, kmer)
74 matches_score(matches, space)
76 # matches.each { |m| puts m.to_s(q_seq) }
78 unless @matches.empty?
79 matches = matches.select { |match| match.score > 0 }
82 if best_match = matches.pop
83 @matches << best_match
85 space_left = Space.new(space.q_min, space.s_min, best_match.q_beg - 1, best_match.s_beg - 1)
86 space_right = Space.new(best_match.q_end + 1, best_match.s_end + 1, space.q_max, space.s_max)
88 align_recurse(q_seq, s_seq, space_left, kmer, matches) unless space_left.empty?
89 align_recurse(q_seq, s_seq, space_right, kmer, matches) unless space_right.empty?
93 # Method to select matches that lies within the search space.
94 def matches_select_by_space(matches, space)
95 new_matches = matches.select do |match|
96 match.q_beg >= space.q_min and
97 match.s_beg >= space.s_min and
98 match.q_end <= space.q_max and
99 match.s_end <= space.s_max
105 def matches_score(matches, space)
106 matches.each do |match|
107 score_length = match_score_length(match)
108 score_diag = match_score_diag(match, space)
110 match.score = score_length + score_diag
113 matches.sort_by! { |match| match.score }
116 def match_score_length(match)
117 match.length * FACTOR_SCORE_LENGTH
120 def match_score_diag(match, space)
121 if space.q_dim >= space.s_dim # s_dim is the narrow end
122 dist_beg = Math.dist_point2line(match.q_beg,
126 space.q_min + space.s_dim,
127 space.s_min + space.s_dim)
129 dist_end = Math.dist_point2line( match.q_beg,
131 space.q_max - space.s_dim,
132 space.s_max - space.s_dim,
136 dist_beg = Math.dist_point2line( match.q_beg,
140 space.q_min + space.q_dim,
141 space.s_min + space.q_dim)
143 dist_end = Math.dist_point2line( match.q_beg,
145 space.q_max - space.q_dim,
146 space.s_max - space.q_dim,
151 dist_min = dist_beg < dist_end ? dist_beg : dist_end
153 dist_min * FACTOR_SCORE_DIAG
156 # Method that finds all maximally expanded non-redundant matches shared
157 # between two sequences inside a given search space.
158 def matches_find(q_seq, s_seq, q_min, s_min, q_max, s_max, kmer)
160 redundant = Hash.new { |h, k| h[k] = [] }
162 s_index = index_seq(s_seq, s_min, s_max, kmer)
166 while q_pos <= q_max - kmer + 1
167 q_oligo = q_seq[q_pos ... q_pos + kmer]
169 s_index[q_oligo].each do |s_pos|
170 match = Match.new(q_pos, s_pos, kmer)
172 unless match_redundant?(redundant, match)
173 match_expand(match, q_seq, s_seq, q_min, s_min, q_max, s_max)
176 match_redundant_add(redundant, match)
186 # Method that indexes a sequence within a given interval such that the
187 # index contains all oligos of a given kmer size and the positions where
188 # this oligo was located.
189 def index_seq(seq, min, max, kmer)
190 index_hash = Hash.new { |h, k| h[k] = [] }
194 while pos <= max - kmer + 1
195 oligo = seq[pos ... pos + kmer]
196 index_hash[oligo] << pos
204 # Method to check if a match is redundant.
205 def match_redundant?(redundant, match)
206 redundant[match.q_beg].each do |s_interval|
207 if s_interval.include? match.s_beg and s_interval.include? match.s_end
215 # Method that adds a match to the redundancy index.
216 def match_redundant_add(redundant, match)
217 (match.q_beg .. match.q_end).each do |q|
218 redundant[q] << (match.s_beg .. match.s_end)
222 # Method that expands a match as far as possible to the left and right.
223 def match_expand(match, q_seq, s_seq, q_min, s_min, q_max, s_max)
224 match_expand_left(match, q_seq, s_seq, q_min, s_min)
225 match_expand_right(match, q_seq, s_seq, q_max, s_max)
230 # Method that expands a match as far as possible to the left.
231 def match_expand_left(match, q_seq, s_seq, q_min, s_min)
232 while match.q_beg > q_min and
233 match.s_beg > s_min and
234 q_seq[match.q_beg - 1] == s_seq[match.s_beg - 1]
243 # Method that expands a match as far as possible to the right.
244 def match_expand_right(match, q_seq, s_seq, q_max, s_max)
245 while match.q_end < q_max and
246 match.s_end < s_max and
247 q_seq[match.q_end + 1] == s_seq[match.s_end + 1]
254 # Method for debugging purposes that upcase matching sequence while non-matches
255 # sequence is kept in lower case.
257 @matches.each do |match|
258 @q_entry.seq[match.q_beg .. match.q_end] = @q_entry.seq[match.q_beg .. match.q_end].upcase
259 @s_entry.seq[match.s_beg .. match.s_end] = @s_entry.seq[match.s_beg .. match.s_end].upcase
263 # Method that insert gaps in sequences based on a list of matches and thus
264 # creating an alignment.
266 @matches.sort_by! { |m| m.q_beg }
271 match = @matches.first
272 diff = (q_gaps + match.q_beg) - (s_gaps + match.s_beg)
275 @q_entry.seq.insert(0, "-" * diff.abs)
278 @s_entry.seq.insert(0, "-" * diff.abs)
282 @matches[1 .. -1].each do |m|
283 diff = (q_gaps + m.q_beg) - (s_gaps + m.s_beg)
286 @q_entry.seq.insert(m.q_beg + q_gaps, "-" * diff.abs)
289 @s_entry.seq.insert(m.s_beg + s_gaps, "-" * diff.abs)
294 diff = @q_entry.length - @s_entry.length
297 @q_entry.seq << ("-" * diff.abs)
299 @s_entry.seq << ("-" * diff.abs)
304 # Class for containing a search space between two sequences q and s.
306 attr_reader :q_min, :s_min, :q_max, :s_max
308 def initialize(q_min, s_min, q_max, s_max)
324 if @q_max - @q_min > 0 and @s_max - @s_min > 0
332 # Class for containing a match between two sequences q and s.
334 attr_accessor :q_beg, :s_beg, :length, :score
336 def initialize(q_beg, s_beg, length, score = 0.0)
352 s = "q: #{@q_beg} #{q_end} s: #{@s_beg} #{s_end} l: #{@length} s: #{@score}"
353 s << " seq: #{seq[@q_beg .. q_end]}" if seq