1 #ifndef UNIFRACWEIGHTEDCOMMAND_H
2 #define UNIFRACWEIGHTEDCOMMAND_H
5 * unifracweightedcommand.h
8 * Created by Sarah Westcott on 2/9/09.
9 * Copyright 2009 Schloss Lab UMASS Amherst. All rights reserved.
13 #include "command.hpp"
16 #include "progress.hpp"
17 #include "sharedutilities.h"
18 #include "fileoutput.h"
21 class UnifracWeightedCommand : public Command {
24 UnifracWeightedCommand(string);
25 UnifracWeightedCommand();
26 ~UnifracWeightedCommand() {}
28 vector<string> setParameters();
29 string getCommandName() { return "unifrac.weighted"; }
30 string getCommandCategory() { return "Hypothesis Testing"; }
31 string getHelpString();
32 string getCitation() { return "Lozupone CA, Hamady M, Kelley ST, Knight R (2007). Quantitative and qualitative beta diversity measures lead to different insights into factors that structure microbial communities. Appl Environ Microbiol 73: 1576-85. \nhttp://www.mothur.org/wiki/Unifrac.weighted"; }
33 string getDescription() { return "generic tests that describes whether two or more communities have the same structure"; }
36 void help() { m->mothurOut(getHelpString()); }
42 linePair(int i, int j) : start(i), num(j) {}
44 vector<linePair> lines;
49 vector<Tree*> T; //user trees
50 vector<double> utreeScores; //user tree unweighted scores
51 vector<double> WScoreSig; //tree weighted score signifigance when compared to random trees - percentage of random trees with that score or lower.
52 vector<string> groupComb; // AB. AC, BC...
54 string sumFile, outputDir;
55 int iters, numGroups, numComp, counter;
56 vector< vector<double> > rScores; //vector<weighted scores for random trees.> each group comb has an entry
57 vector< vector<double> > uScores; //vector<weighted scores for user trees.> each group comb has an entry
58 vector< map<float, float> > rScoreFreq; //map <weighted score, number of random trees with that score.> -vector entry for each combination.
59 vector< map<float, float> > rCumul; //map <weighted score, cumulative percentage of number of random trees with that score or higher.> -vector entry for each c
60 map<float, float> validScores; //map contains scores from random
62 bool abort, phylip, random, includeRoot, subsample, consensus;
63 string groups, itersString, outputForm, treefile, groupfile, namefile;
64 vector<string> Groups, outputNames; //holds groups to be used
65 int processors, numUniquesInName, subsampleSize, subsampleIters;
67 map<string, string> nameMap;
69 void printWSummaryFile();
70 void printWeightedFile();
71 void createPhylipFile();
72 //void removeValidScoresDuplicates();
73 int findIndex(float, int);
74 void calculateFreqsCumuls();
75 int createProcesses(Tree*, vector< vector<string> >, vector< vector<double> >&);
76 int driver(Tree*, vector< vector<string> >, int, int, vector< vector<double> >&);
78 int runRandomCalcs(Tree*, vector<double>);
80 vector<Tree*> buildTrees(vector< vector<double> >&, int, TreeMap*);
81 int getConsensusTrees(vector< vector<double> >&, int);
82 int getAverageSTDMatrices(vector< vector<double> >&, int);