# >>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>><<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<
-# Class containing methods for located all Maximally Extended Matches (MEMs) between
-# two strings.
+require 'profile'
+
+# Class containing methods for located all non-redundant Maximally Extended Matches (MEMs)
+# between two strings.
class Matches
- # Class method to located all MEMs bewteen two sequences within a given search
- # space and seed with kmers of a given size.
+ # Class method to located all non-redudant MEMs bewteen two sequences within a given
+ # search space and seeded with kmers of a given size.
def self.find(q_seq, s_seq, q_min, s_min, q_max, s_max, kmer)
m = self.new
m.matches_find(q_seq, s_seq, q_min, s_min, q_max, s_max, kmer)
s_index[q_oligo].each do |s_pos|
match = Match.new(q_pos, s_pos, kmer)
+
unless match_redundant?(redundant, match)
match_expand(match, q_seq, s_seq, q_min, s_min, q_max, s_max)
matches << match
# Method to check if a match is redundant.
def match_redundant?(redundant, match)
redundant[match.q_beg].each do |s_interval|
- if s_interval.include? match.s_beg and s_interval.include? match.s_end
+ if s_interval.include? match.s_beg and s_interval.include? match.s_end # TODO test if include? is slow
return true
end
end
FACTOR_SCORE_LENGTH = 1.0
FACTOR_SCORE_DIAG = -1.41
+KMER = 32
# Module with stuff to create a pairwise aligment.
module PairAlign
@s_entry.seq.downcase!
space = Space.new(0, 0, @q_entry.length - 1, @s_entry.length - 1)
- kmer = 32
- align_recurse(@q_entry.seq, @s_entry.seq, space, kmer)
+ align_recurse(@q_entry.seq, @s_entry.seq, space, KMER)
matches_upcase
gaps_insert
end
dist_min * FACTOR_SCORE_DIAG
end
- # Method that finds all maximally expanded non-redundant matches shared
- # between two sequences inside a given search space.
- def matches_find(q_seq, s_seq, q_min, s_min, q_max, s_max, kmer)
- matches = []
- redundant = Hash.new { |h, k| h[k] = [] }
-
- s_index = index_seq(s_seq, s_min, s_max, kmer)
-
- q_pos = q_min
-
- while q_pos <= q_max - kmer + 1
- q_oligo = q_seq[q_pos ... q_pos + kmer]
-
- s_index[q_oligo].each do |s_pos|
- match = Match.new(q_pos, s_pos, kmer)
-
- unless match_redundant?(redundant, match)
- match_expand(match, q_seq, s_seq, q_min, s_min, q_max, s_max)
- matches << match
-
- match_redundant_add(redundant, match)
- end
- end
-
- q_pos += 1
- end
-
- matches
- end
-
- # Method that indexes a sequence within a given interval such that the
- # index contains all oligos of a given kmer size and the positions where
- # this oligo was located.
- def index_seq(seq, min, max, kmer)
- index_hash = Hash.new { |h, k| h[k] = [] }
-
- pos = min
-
- while pos <= max - kmer + 1
- oligo = seq[pos ... pos + kmer]
- index_hash[oligo] << pos
-
- pos += 1
- end
-
- index_hash
- end
-
- # Method to check if a match is redundant.
- def match_redundant?(redundant, match)
- redundant[match.q_beg].each do |s_interval|
- if s_interval.include? match.s_beg and s_interval.include? match.s_end
- return true
- end
- end
-
- false
- end
-
- # Method that adds a match to the redundancy index.
- def match_redundant_add(redundant, match)
- (match.q_beg .. match.q_end).each do |q|
- redundant[q] << (match.s_beg .. match.s_end)
- end
- end
-
- # Method that expands a match as far as possible to the left and right.
- def match_expand(match, q_seq, s_seq, q_min, s_min, q_max, s_max)
- match_expand_left(match, q_seq, s_seq, q_min, s_min)
- match_expand_right(match, q_seq, s_seq, q_max, s_max)
-
- match
- end
-
- # Method that expands a match as far as possible to the left.
- def match_expand_left(match, q_seq, s_seq, q_min, s_min)
- while match.q_beg > q_min and
- match.s_beg > s_min and
- q_seq[match.q_beg - 1] == s_seq[match.s_beg - 1]
- match.q_beg -= 1
- match.s_beg -= 1
- match.length += 1
- end
-
- match
- end
-
- # Method that expands a match as far as possible to the right.
- def match_expand_right(match, q_seq, s_seq, q_max, s_max)
- while match.q_end < q_max and
- match.s_end < s_max and
- q_seq[match.q_end + 1] == s_seq[match.s_end + 1]
- match.length += 1
- end
-
- match
- end
-
# Method for debugging purposes that upcase matching sequence while non-matches
# sequence is kept in lower case.
def matches_upcase
true
end
end
-
- # Class for containing a match between two sequences q and s.
- class Match
- attr_accessor :q_beg, :s_beg, :length, :score
-
- def initialize(q_beg, s_beg, length, score = 0.0)
- @q_beg = q_beg
- @s_beg = s_beg
- @length = length
- @score = score
- end
-
- def q_end
- @q_beg + @length - 1
- end
-
- def s_end
- @s_beg + @length - 1
- end
-
- def to_s(seq = nil)
- s = "q: #{@q_beg} #{q_end} s: #{@s_beg} #{s_end} l: #{@length} s: #{@score}"
- s << " seq: #{seq[@q_beg .. q_end]}" if seq
- s
- end
- end
end