3588a32896eff48854b2e16d813f5693a4fbf045
braney
  Tue Jun 13 15:35:58 2017 -0700
remove unneeded error message from optimalLeaf library

diff --git src/optimalLeaf/SimMat.cc src/optimalLeaf/SimMat.cc
index 370f7a7..51c4457 100644
--- src/optimalLeaf/SimMat.cc
+++ src/optimalLeaf/SimMat.cc
@@ -1,225 +1,223 @@
 /* This code provided by Ziv Bar-Joseph with the explicit understanding that 
  * no licensing is required for it to be used in the UCSC Genome Browser
  * as long as reference is made to this paper:
  * https://www.ncbi.nlm.nih.gov/pubmed/12801867
  */
 
 // --------------------------------------------------------
 //                     SimMat.cc
 // Computes the similarity matrix, and writes ordered trees 
 // to a .CDT files.
 // --------------------------------------------------------
 #include <stdlib.h>
 #include <iostream>
 #include <fstream>
 #include <string.h>
 #include <math.h>
 #include "SimMat.hh"
 #include "readFile.hh"
 
 using namespace std;
 // read the input file
 SimMat::SimMat(char *filename,int orfL,int descL,int mZ) {
 
   meanZ = mZ;
   readFile myRead(filename,orfL,descL);
   genes = myRead.getNumGenes();
   exp = myRead.getNumExp();
   in = myRead.getVals();
   geneNames = myRead.getNames();
   geneDesc = myRead.getDesc();
   expNames = myRead.getExp();
   generateMat();
 }
 
 SimMat::SimMat( int igenes, int iexp, int imeanZ, float **iin,
   char **igeneNames, char **igeneDesc, char **iexpNames) {
 
   genes = igenes;
   exp = iexp;
   meanZ = imeanZ;
   in = iin;
   geneNames = igeneNames;
   geneDesc = igeneDesc;
   expNames = iexpNames;
 
   generateMat();
 }
 
 // compute the siimlarity matrix
 void SimMat::generateMat() {
   
   int i,j;
   mat = new float*[genes+1];
   means = compMean(genes,exp,1);
   std = compStd(genes,exp,1);
   for(i=1;i<genes+1;i++) {
     mat[i] = new float[genes+1];
     mat[i][i] = 0;
   }
   for(i=1;i<genes;i++) {
     for(j=i+1;j<genes+1;j++) {
       mat[i][j] = -1 * compSim(i-1,j-1,exp,1);
       mat[j][i] = mat[i][j];
     }
   } 
 }
 
 // compute the columns siimlarity matrix
 void SimMat::generateCols(char *name) {
   
   int i,j;
   delete []means;
   delete []std;
   float simVal;
   ofstream toSim(name);
   cMat = new float*[exp+1];
   means = compMean(exp,genes,0);
   std = compStd(exp,genes,0);
   for(i=1;i<exp+1;i++) {
     cMat[i] = new float[exp+1];
     cMat[i][i] = 0;
   }
   for(i=1;i<exp;i++) {
     for(j=i+1;j<exp+1;j++) {
       simVal = compSim(i-1,j-1,genes,0);
       cMat[i][j] = -1 * simVal;
       cMat[j][i] = cMat[i][j];
     }
   } 
 }
 // computes average expressions of genes
 float *SimMat::compMean(int tot,int c,int isGene) {
   int i,j;
   float *res = new float[tot];
   float sum;
   int num;
   for(i=0;i<tot;i++) {
     sum = 0;
     num = 0;
     for(j=0;j<c;j++) {
       if(isGene == 1) {
 	if(in[i][j] < 100) {
 	  sum += in[i][j];
 	  num++;
 	}
       }
       else {
         if(in[j][i] < 100) {
 	  sum += in[j][i];
 	  num++;
 	}
       }
     }
     res[i] = sum/(float)num;
     if(meanZ)
       res[i] = 0;
   }
   return res;
 }
 // computes standard devaition of genes
 float *SimMat::compStd(int tot,int c, int isGene) {
 
   int i,j;
   float *res = new float[tot];
   float sum;
   int num;
   for(i=0;i<tot;i++) {
     sum = 0;
     num = 0;
     for(j=0;j<c;j++) {
       if(isGene == 1) {
 	if(in[i][j] < 100) {
 	  sum += pow((in[i][j]-means[i]),2);
 	  num++;
 	}
       }
       else {
         if(in[j][i] < 100) {
 	  sum += pow((in[j][i]-means[i]),2);
 	  num++;
 	}
       }
       
     }
     sum = sum/(float)num;
     if(sum == 0) {
-      cerr<<"Error! no variance for gene "<<i<<'\n';
-      cerr<<"The variance for gene "<<i<<" has been set to 1"<<'\n';
       sum = 1;
     }
     res[i] = sqrt(sum);
   }
   return res;
 }
 
 float SimMat::compSim(int g1,int g2,int c,int isGene) {
   int i;
   float sum = 0;
   int num = 0;
   for(i=0;i<c;i++) {
     if(isGene == 1) {
       if(in[g1][i] < 100 && in[g2][i] < 100) {
 	sum += ((in[g1][i]-means[g1])/std[g1])*((in[g2][i]-means[g2])/std[g2]);
 	num++;
       }
     }
     else {
       if(in[i][g1] < 100 && in[i][g2] < 100) {
 	sum += ((in[i][g1]-means[g1])/std[g1])*((in[i][g2]-means[g2])/std[g2]);
 	num++;
       }
     }   
   }
   if(num == 0) 
     sum = -1;
   else
     sum = sum/(float)num;
   return sum;
 }
 
 // gets an array representing genes, and writes it to a .CDT file 
 void SimMat::printOrder(int *arr,int *arrC,char *name) {
 
   ofstream to(name);
   int i,j;
   to<<"GID"<<'\t'<<"gene"<<'\t'<<"NAME"<<'\t'<<"GWEIGHT";
   for(i=0;i<exp;i++) {
     to<<'\t'<<expNames[arrC[i]-1]<<"-"<<arrC[i];
   }
   to<<'\n';
   to<<"AID"<<'\t'<<'\t'<<'\t';
   for(i=0;i<exp;i++) {
     to<<'\t'<<"ARRY"<<arrC[i]<<"X";
   }
   to<<'\n';
   to<<"EWEIGHT"<<'\t'<<'\t'<<'\t';
   for(i=1;i<exp+1;i++) {
     to<<'\t'<<"1";
   }
   to<<'\n';
   for(i=0;i<genes;i++) {
     to<<"GENE"<<arr[i]<<"X"<<'\t'<<geneNames[arr[i]-1]<<'\t'<<geneDesc[arr[i]-1]<<'\t'<<"1";
     for(j=0;j<exp;j++) {
       if(in[arr[i]-1][arrC[j]-1] < 100)
 	to<<'\t'<<in[arr[i]-1][arrC[j]-1];
       else
 	to<<'\t';
     }
     to<<'\n';
   }
 }