1 /***************************************************************************
2 * Copyright (C) 2006 by Michael Andreen *
3 * andreen@student.chalmers.se *
5 * This program is free software; you can redistribute it and/or modify *
6 * it under the terms of the GNU General Public License as published by *
7 * the Free Software Foundation; either version 2 of the License, or *
8 * (at your option) any later version. *
10 * This program is distributed in the hope that it will be useful, *
11 * but WITHOUT ANY WARRANTY; without even the implied warranty of *
12 * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the *
13 * GNU General Public License for more details. *
15 * You should have received a copy of the GNU General Public License *
16 * along with this program; if not, write to the *
17 * Free Software Foundation, Inc., *
18 * 51 Franklin Street, Fifth Floor, Boston, MA 02110-1301, USA *
19 ***************************************************************************/
21 #include "genealgorithms.h"
22 #include "geneorder.h"
23 #include "componenttree.h"
32 std::pair<int,int> longestSequences(const GeneOrder& go){
33 vector<vector<int> > v = robinsonSchensted(go);
34 return pair<int,int>(v[0].size(),v.size());
37 std::vector<std::vector<int> > robinsonSchensted(const GeneOrder& go){
38 vector<vector<int> > v;
39 for (GeneOrder::iterator i = go.begin(); i != go.end(); ++i){
42 for (vector<vector<int> >::iterator vs = v.begin();
44 vector<int>::iterator bigger = upper_bound(vs->begin(),vs->end(),n);
45 if ( bigger == vs->end()){
54 v.push_back(vector<int>());
55 v.back().push_back(n);
63 FindP(size_t p) : p(p) {}
64 bool operator()(Interval i){
65 return (i.first == p || i.second == p);
70 size_t countCycles(const GeneOrder& go){
73 vector<Interval> intervals = findIntervals(go);
74 vector<Interval> points = findIntervalsAtPoints(intervals);
75 for (size_t p = 1; p < go.size(); ++p){
76 if (marked.find(p) != marked.end())
78 Interval i = intervals[points[p].first];
79 while (marked.find(p) == marked.end()){
81 if (i == intervals[points[p].first])
82 i = intervals[points[p].second];
84 i = intervals[points[p].first];
96 size_t inversionDistance(const GeneOrder& go){
97 size_t cycles = countCycles(go);
99 return go.size() - 1 - cycles;
111 Gene operator()(Gene x) const{
115 std::vector<Component> findComponents(const GeneOrder& go){
116 vector<Component> components;
117 vector<int> os(go.size()-1);
118 for (size_t i = 0; i < os.size(); ++i)
119 os[i] = (go[i]*go[i+1] > 0 ? sign(go[i]) : 0);
121 Mdir.push(go.size()-1);
129 dir.push_back(go.size()-1);
133 vector<Gene> p(go.list());
134 transform(p.begin(),p.end(),p.begin(),Abs());
135 for (size_t i = 1; i < go.size(); ++i){
139 else while (Mdir.top() < p[i])
141 dir.push_back(Mdir.top());
144 while(p[Sdir.top()] > p[i] || dir[Sdir.top()] < p[i]){
146 os[Sdir.top()] = (os[Sdir.top()] == os[s] ? os[s] : 0);
149 if (go[i] > 0 && dir[i] == dir[s] && static_cast<Gene>(i - s) == p[i] - p[s])
150 components.push_back(Component(p[s],p[i],(s+1 == i ? 0 : os[s]),s,i));
155 else while (Mrev.top() > p[i])
157 rev.push_back(Mrev.top());
160 while((p[s] < p[i] || rev[s] > p[i]) && s > 0){
162 os[Srev.top()] *= (os[Srev.top()] == os[s] ? 1 : 0);
165 if (go[i] < 0 && rev[i] == rev[s] && static_cast<Gene>(i - s) == p[s] - p[i])
166 components.push_back(Component(-p[s],-p[i],(s+1 == i ? 0 : os[s]),s,i));
185 std::vector<Interval> findIntervals(const GeneOrder& go){
186 const size_t max = go.size();
187 vector<Interval> intervals(go.size()-1,Interval(max,max,false));
189 const GeneOrder::iterator end = go.end();
190 for (GeneOrder::iterator g = go.begin(); g != end; ++g, ++n){
193 Interval& curr = intervals[i];
194 curr.first = n + (*g >= 0 ? 1 : 0);
196 if (curr.second == max)
197 curr.oriented = *g < 0;
199 curr.oriented ^= *g < 0;
202 Interval& prev = intervals[i-1];
203 prev.second = n + (*g < 0 ? 1 : 0);
205 if (prev.first == max)
206 prev.oriented = *g < 0;
208 prev.oriented ^= *g < 0;
218 std::vector<Interval> findIntervalsAtPoints(const vector<Interval>& intervals){
219 size_t max = intervals.size()+1;
220 vector<Interval> points(max,Interval(max,max,false));
222 for (vector<Interval>::const_iterator i = intervals.begin(); i != intervals.end(); ++i, ++n){
223 if (points[i->first].first == max){
224 points[i->first].first = n;
226 points[i->first].second = n;
228 if (points[i->second].first == max){
229 points[i->second].first = n;
231 points[i->second].second = n;