5 * Created by westcott on 11/20/09.
6 * Copyright 2009 Schloss Lab. All rights reserved.
10 #include "phylotypecommand.h"
11 #include "phylotree.h"
12 #include "listvector.hpp"
13 #include "rabundvector.hpp"
14 #include "sabundvector.hpp"
16 /**********************************************************************************************************************/
17 PhylotypeCommand::PhylotypeCommand(string option) {
21 //allow user to run help
22 if(option == "help") { help(); abort = true; }
26 //valid paramters for this command
27 string AlignArray[] = {"taxonomy","cutoff","label","name","outputdir","inputdir"};
28 vector<string> myArray (AlignArray, AlignArray+(sizeof(AlignArray)/sizeof(string)));
30 OptionParser parser(option);
31 map<string, string> parameters = parser.getParameters();
33 ValidParameters validParameter;
34 map<string, string>::iterator it;
36 //check to make sure all parameters are valid for command
37 for (it = parameters.begin(); it != parameters.end(); it++) {
38 if (validParameter.isValidParameter(it->first, myArray, it->second) != true) { abort = true; }
41 //if the user changes the input directory command factory will send this info to us in the output parameter
42 string inputDir = validParameter.validFile(parameters, "inputdir", false);
43 if (inputDir == "not found"){ inputDir = ""; }
46 it = parameters.find("taxonomy");
47 //user has given a template file
48 if(it != parameters.end()){
49 path = hasPath(it->second);
50 //if the user has not given a path then, add inputdir. else leave path alone.
51 if (path == "") { parameters["taxonomy"] = inputDir + it->second; }
54 it = parameters.find("name");
55 //user has given a template file
56 if(it != parameters.end()){
57 path = hasPath(it->second);
58 //if the user has not given a path then, add inputdir. else leave path alone.
59 if (path == "") { parameters["name"] = inputDir + it->second; }
63 taxonomyFileName = validParameter.validFile(parameters, "taxonomy", true);
64 if (taxonomyFileName == "not found") {
65 m->mothurOut("taxonomy is a required parameter for the phylotype command.");
66 m->mothurOutEndLine();
68 }else if (taxonomyFileName == "not open") { abort = true; }
70 namefile = validParameter.validFile(parameters, "name", true);
71 if (namefile == "not open") { abort = true; }
72 else if (namefile == "not found") { namefile = ""; }
73 else { readNamesFile(); }
75 //if the user changes the output directory command factory will send this info to us in the output parameter
76 outputDir = validParameter.validFile(parameters, "outputdir", false); if (outputDir == "not found"){
78 outputDir += hasPath(taxonomyFileName); //if user entered a file with a path then preserve it
81 string temp = validParameter.validFile(parameters, "cutoff", false);
82 if (temp == "not found") { temp = "-1"; }
83 convert(temp, cutoff);
85 label = validParameter.validFile(parameters, "label", false);
86 if (label == "not found") { label = ""; allLines = 1; }
88 if(label != "all") { splitAtDash(label, labels); allLines = 0; }
89 else { allLines = 1; }
95 m->errorOut(e, "PhylotypeCommand", "PhylotypeCommand");
99 /**********************************************************************************************************************/
101 void PhylotypeCommand::help(){
103 m->mothurOut("The phylotype command reads a taxonomy file and outputs a .list, .rabund and .sabund file. \n");
104 m->mothurOut("The phylotype command parameter options are taxonomy, cutoff and label. The taxonomy parameter is required.\n");
105 m->mothurOut("The cutoff parameter allows you to specify the level you want to stop at. The default is the highest level in your taxonomy file. \n");
106 m->mothurOut("For example: taxonomy = Bacteria;Bacteroidetes-Chlorobi;Bacteroidetes; - cutoff=2, would truncate the taxonomy to Bacteria;Bacteroidetes-Chlorobi; \n");
107 m->mothurOut("For the cutoff parameter levels count up from the root of the phylotree. This enables you to look at the grouping down to a specific resolution, say the genus level.\n");
108 m->mothurOut("The label parameter allows you to specify which level you would like, and are separated by dashes. The default all levels in your taxonomy file. \n");
109 m->mothurOut("For the label parameter, levels count down from the root to keep the output similiar to mothur's other commands which report information from finer resolution to coarser resolutions.\n");
110 m->mothurOut("The phylotype command should be in the following format: \n");
111 m->mothurOut("phylotype(taxonomy=yourTaxonomyFile, cutoff=yourCutoff, label=yourLabels) \n");
112 m->mothurOut("Eaxample: phylotype(taxonomy=amazon.taxonomy, cutoff=5, label=1-3-5).\n\n");
114 catch(exception& e) {
115 m->errorOut(e, "PhylotypeCommand", "help");
119 /**********************************************************************************************************************/
121 PhylotypeCommand::~PhylotypeCommand(){}
123 /**********************************************************************************************************************/
125 int PhylotypeCommand::execute(){
128 if (abort == true) { return 0; }
130 vector<string> outputNames;
132 //reads in taxonomy file and makes all the taxonomies the same length
133 //by appending the last taxon to a given taxonomy as many times as needed to
134 //make it as long as the longest taxonomy in the file
135 TaxEqualizer* taxEqual = new TaxEqualizer(taxonomyFileName, cutoff, outputDir);
137 if (m->control_pressed) { delete taxEqual; return 0; }
139 string equalizedTaxFile = taxEqual->getEqualizedTaxFile();
143 //build taxonomy tree from equalized file
144 PhyloTree* tree = new PhyloTree(equalizedTaxFile);
145 vector<int> leaves = tree->getGenusNodes();
147 //store leaf nodes in current map
148 for (int i = 0; i < leaves.size(); i++) { currentNodes[leaves[i]] = leaves[i]; }
151 if (tree->get(leaves[0]).parent == -1) { m->mothurOut("Empty Tree"); m->mothurOutEndLine(); done = true; }
153 if (m->control_pressed) { delete tree; return 0; }
155 string fileroot = outputDir + getRootName(getSimpleName(taxonomyFileName));
158 string outputListFile = fileroot + "tx.list";
159 openOutputFile(outputListFile, outList);
161 string outputSabundFile = fileroot + "tx.sabund";
162 openOutputFile(outputSabundFile, outSabund);
164 string outputRabundFile = fileroot + "tx.rabund";
165 openOutputFile(outputRabundFile, outRabund);
167 outputNames.push_back(outputListFile);
168 outputNames.push_back(outputSabundFile);
169 outputNames.push_back(outputRabundFile);
172 //start at leaves of tree and work towards root, processing the labels the user wants
173 while((!done) && ((allLines == 1) || (labels.size() != 0))) {
175 string level = toString(count);
178 if (m->control_pressed) {
179 outRabund.close(); outSabund.close(); outList.close();
180 for (int i = 0; i < outputNames.size(); i++) { remove(outputNames[i].c_str()); }
181 delete tree; return 0;
184 //is this a level the user want output for
185 if(allLines == 1 || labels.count(level) == 1){
188 m->mothurOut(level); m->mothurOutEndLine();
191 list.setLabel(level);
192 //go through nodes and build listvector
193 for (itCurrent = currentNodes.begin(); itCurrent != currentNodes.end(); itCurrent++) {
196 TaxNode node = tree->get(itCurrent->first);
197 parentNodes[node.parent] = node.parent;
199 vector<string> names = node.accessions;
201 //make the names compatable with listvector
203 for (int i = 0; i < names.size(); i++) {
204 if (namefile != "") {
205 map<string, string>::iterator itNames = namemap.find(names[i]); //make sure this name is in namefile
207 if (itNames != namemap.end()) { name += namemap[names[i]] + ","; } //you found it in namefile
208 else { m->mothurOut(names[i] + " is not in your namefile, please correct."); m->mothurOutEndLine(); exit(1); }
210 }else{ name += names[i] + ","; }
212 name = name.substr(0, name.length()-1); //rip off extra ','
214 //add bin to list vector
215 list.push_back(name);
221 list.getRAbundVector().print(outRabund);
223 list.getSAbundVector().print(outSabund);
230 for (itCurrent = currentNodes.begin(); itCurrent != currentNodes.end(); itCurrent++) {
231 int parent = tree->get(itCurrent->first).parent;
232 parentNodes[parent] = parent;
237 currentNodes = parentNodes;
240 //have we reached the rootnode
241 if (tree->get(currentNodes.begin()->first).parent == -1) { done = true; }
250 if (m->control_pressed) {
251 for (int i = 0; i < outputNames.size(); i++) { remove(outputNames[i].c_str()); }
255 m->mothurOutEndLine();
256 m->mothurOut("Output File Names: "); m->mothurOutEndLine();
257 for (int i = 0; i < outputNames.size(); i++) { m->mothurOut(outputNames[i]); m->mothurOutEndLine(); }
258 m->mothurOutEndLine();
263 catch(exception& e) {
264 m->errorOut(e, "PhylotypeCommand", "execute");
268 /*****************************************************************/
269 int PhylotypeCommand::readNamesFile() {
273 openInputFile(namefile, in);
275 string first, second;
276 map<string, string>::iterator itNames;
279 in >> first >> second; gobble(in);
281 itNames = namemap.find(first);
282 if (itNames == namemap.end()) {
283 namemap[first] = second;
284 }else { m->mothurOut(first + " has already been seen in namefile, disregarding names file."); m->mothurOutEndLine(); in.close(); namemap.clear(); namefile = ""; return 1; }
290 catch(exception& e) {
291 m->errorOut(e, "PhylotypeCommand", "readNamesFile");
296 /**********************************************************************************************************************/