#include "knn.h"
/**************************************************************************************************/
-Knn::Knn(string tfile, string tempFile, string method, int kmerSize, int gapOpen, int gapExtend, int match, int misMatch, int n)
-: Classify(tfile, tempFile, method, kmerSize, gapOpen, gapExtend, match, misMatch), num(n) {}
+Knn::Knn(string tfile, string tempFile, string method, int kmerSize, float gapOpen, float gapExtend, float match, float misMatch, int n)
+: Classify(), num(n), search(method) {
+ try {
+ //create search database and names vector
+ generateDatabaseAndNames(tfile, tempFile, method, kmerSize, gapOpen, gapExtend, match, misMatch);
+ }
+ catch(exception& e) {
+ m->errorOut(e, "Knn", "Knn");
+ exit(1);
+ }
+}
+/**************************************************************************************************/
+void Knn::setDistName(string s) {
+ try {
+ outDistName = s;
+ ofstream outDistance;
+ m->openOutputFile(outDistName, outDistance);
+ outDistance << "Name\tBestMatch\tDistance" << endl;
+ outDistance.close();
+ }
+ catch(exception& e) {
+ m->errorOut(e, "Knn", "setDistName");
+ exit(1);
+ }
+}
+/**************************************************************************************************/
+Knn::~Knn() {
+ try {
+ delete phyloTree;
+ if (database != NULL) { delete database; }
+ }
+ catch(exception& e) {
+ m->errorOut(e, "Knn", "~Knn");
+ exit(1);
+ }
+}
/**************************************************************************************************/
string Knn::getTaxonomy(Sequence* seq) {
try {
//use database to find closest seq
vector<int> closest = database->findClosestSequences(seq, num);
-
+
+ if (search == "distance") { ofstream outDistance; m->openOutputFileAppend(outDistName, outDistance); outDistance << seq->getName() << '\t' << database->getName(closest[0]) << '\t' << database->getSearchScore() << endl; outDistance.close(); }
+
+ if (m->control_pressed) { return tax; }
+
vector<string> closestNames;
for (int i = 0; i < closest.size(); i++) {
//find that sequences taxonomy in map
//is this sequence in the taxonomy file
if (it == taxonomy.end()) { //error not in file
- mothurOut("Error: sequence " + names[closest[i]] + " is not in the taxonomy file. It will be eliminated as a match to sequence " + seq->getName() + "."); mothurOutEndLine();
+ m->mothurOut("Error: sequence " + names[closest[i]] + " is not in the taxonomy file. It will be eliminated as a match to sequence " + seq->getName() + "."); m->mothurOutEndLine();
}else{ closestNames.push_back(it->first); }
}
if (closestNames.size() == 0) {
- mothurOut("Error: All the matches for sequence " + seq->getName() + " have been eliminated. " + seq->getName() + " will be disregarded."); mothurOutEndLine();
+ m->mothurOut("Error: All the matches for sequence " + seq->getName() + " have been eliminated. " + seq->getName() + " will be disregarded."); m->mothurOutEndLine();
tax = "bad seq";
}else{
tax = findCommonTaxonomy(closestNames);
- if (tax == "") { mothurOut("There are no common levels for sequence " + seq->getName() + ". " + seq->getName() + " will be disregarded."); mothurOutEndLine(); tax = "bad seq"; }
+ if (tax == "") { m->mothurOut("There are no common levels for sequence " + seq->getName() + ". " + seq->getName() + " will be disregarded."); m->mothurOutEndLine(); tax = "bad seq"; }
}
+ simpleTax = tax;
return tax;
}
catch(exception& e) {
- errorOut(e, "Knn", "getTaxonomy");
+ m->errorOut(e, "Knn", "getTaxonomy");
exit(1);
}
}
int smallest = 100;
for (int i = 0; i < closest.size(); i++) {
+ if (m->control_pressed) { return "control"; }
string tax = taxonomy[closest[i]]; //we know its there since we checked in getTaxonomy
//start at the highest level all the closest seqs have
string common = "";
for (int i = (smallest-1); i >= 0; i--) {
+ if (m->control_pressed) { return "control"; }
string thistax = taxons[0][i];
int num = 0;
return common;
}
catch(exception& e) {
- errorOut(e, "Knn", "findCommonTaxonomy");
+ m->errorOut(e, "Knn", "findCommonTaxonomy");
exit(1);
}
}