X-Git-Url: https://git.donarmstrong.com/?p=mothur.git;a=blobdiff_plain;f=ccode.h;h=456b735df524ff3eed9c2688be95f8537c0d0c86;hp=a4860dcb83ee16ddaca43afd6cc4adb4a1973adc;hb=a8e2df1b96a57f5f29576b08361b86a96a8eff4f;hpb=f7fbd74ffedcf62217109c22e828453eaefa1458

diff --git a/ccode.h b/ccode.h
index a4860dc..456b735 100644
--- a/ccode.h
+++ b/ccode.h
@@ -14,6 +14,7 @@
 #include "dist.h"
 #include "decalc.h"
 
+/***********************************************************/
 //This class was created using the algorythms described in the 
 // "Evaluating putative chimeric sequences from PCR-amplified products" paper 
 //by Juan M. Gonzalez, Johannes Zimmerman and Cesareo Saiz-Jimenez.
@@ -23,36 +24,66 @@
 class Ccode : public Chimera {
 	
 	public:
-		Ccode(string, string);	
+		Ccode(string, string, bool, string, int, int, string);	//fasta, template, filter, mask, window, numWanted, outputDir
 		~Ccode();
 		
-		void getChimeras();
-		void print(ostream&);
-		
-		void setCons(string c)		{}
-		void setQuantiles(string q) {}
-		
+		int getChimeras(Sequence* query);
+		Sequence print(ostream&, ostream&);
 		
+		#ifdef USE_MPI
+		Sequence print(MPI_File&, MPI_File&);
+		#endif
 	private:
 	
 		Dist* distCalc;
 		DeCalculator* decalc;
-		int iters;
-		string fastafile, templateFile;
+		int iters, window, numWanted;
+		string fastafile, mapInfo;
 		
+		Sequence* querySeq;
 		
-		vector<linePair*> lines;
-		vector<linePair*> templateLines;
-		vector<Sequence*> querySeqs;
-		vector<Sequence*> templateSeqs;
+		map<int, int> spotMap;
+		map<int, int>::iterator it;
 		
-		vector< vector<Sequence*> > closest;  //bestfit[0] is a vector of sequence at are closest to queryseqs[0]...
+		vector<int>  windows; //windows is the vector of window breaks for query
+		int windowSizes;  //windowSizes is the size of the windows for query
+		map<int, int> trim;  //trim is the map containing the starting and ending positions for query
+		vector<SeqDist>  closest;  //closest is a vector of sequence at are closest to query
+		vector<float>  averageRef;  //averageRef is the average distance at each window for the references for query
+		vector<float>  averageQuery;  //averageQuery is the average distance at each winow for the query for query
+		vector<float>   sumRef;  //sumRef is the sum of distances at each window for the references for query
+		vector<float>   sumSquaredRef;  //sumSquaredRef is the sum of squared distances at each window for the references for query
+		vector<float> sumQuery;  //sumQuery is the sum of distances at each window for the comparison of query to references for query
+		vector<float>  sumSquaredQuery;  //sumSquaredQuery is the sum of squared distances at each window for the comparison of query to references for query
+		vector<float> varRef;  //varRef is the variance among references seqs at each window for query
+		vector<float> varQuery;  //varQuery is the variance among references and query at each window
+		vector<float> sdRef;  //sdRef is the standard deviation of references seqs at each window for query
+		vector<float> sdQuery;  //sdQuery is the standard deviation of references and query at each window
+		vector<float> anova;  //anova is the vector of anova scores for each window for query
+		int refCombo;  //refCombo is the number of reference sequences combinations for query
+		vector<bool>  isChimericConfidence;  //isChimericConfidence indicates whether query is chimeric at a given window according to the confidence limits
+		vector<bool>  isChimericTStudent;  //isChimericConfidence indicates whether query is chimeric at a given window according to the confidence limits
+		vector<bool>  isChimericANOVA;  //isChimericConfidence indicates whether query is chimeric at a given window according to the confidence limits
 		
-		vector< vector<Sequence*> > findClosest(int, int, int); 
-		void removeSeqs(vector<Sequence*>);  //removes sequences from closest that are to different of too similar to eachother. 
+		vector<SeqDist>  findClosest(Sequence*, int); 
+		void removeBadReferenceSeqs(vector<SeqDist>&);  //removes sequences from closest that are to different of too similar to eachother. 
+		void trimSequences(Sequence*);
+		vector<int> findWindows();
+		void getAverageRef(vector<SeqDist>);		//fills sumRef, averageRef, sumSquaredRef and refCombo.
+		void getAverageQuery (vector<SeqDist>, Sequence*);	//fills sumQuery, averageQuery, sumSquaredQuery.
+		void findVarianceRef ();						//fills varRef and sdRef also sets minimum error rate to 0.001 to avoid divide by 0.
+		void findVarianceQuery ();					//fills varQuery and sdQuery
+		void determineChimeras ();					//fills anova, isChimericConfidence, isChimericTStudent and isChimericANOVA.
 		
-		void createProcessesClosest();
+		int getDiff(string, string);  //return number of mismatched bases, a gap to base is not counted as a mismatch
+		float getT(int); 
+		float getF(int); 
 		
+		#ifdef USE_MPI
+		int printMapping(string&);
+		MPI_File outMap;
+		#endif
+
 };
 
 /***********************************************************/