From ecd4f324189cf1cb49a3eb49d5cff58562f27b7f Mon Sep 17 00:00:00 2001 From: Sarah Westcott Date: Wed, 1 Feb 2012 13:13:47 -0500 Subject: [PATCH] added significance scores to core.axes. added design file to get.groups and remove.groups --- corraxescommand.cpp | 39 ++++++++++++++++-- getgroupscommand.cpp | 90 +++++++++++++++++++++++++++++++++++++---- getgroupscommand.h | 5 ++- removegroupscommand.cpp | 84 +++++++++++++++++++++++++++++++++++--- removegroupscommand.h | 5 ++- 5 files changed, 204 insertions(+), 19 deletions(-) diff --git a/corraxescommand.cpp b/corraxescommand.cpp index 4e9c3c6..10669e5 100644 --- a/corraxescommand.cpp +++ b/corraxescommand.cpp @@ -275,7 +275,7 @@ int CorrAxesCommand::execute(){ if (metadatafile == "") { out << "OTU"; } else { out << "Feature"; } - for (int i = 0; i < numaxes; i++) { out << '\t' << "axis" << (i+1); } + for (int i = 0; i < numaxes; i++) { out << '\t' << "axis" << (i+1) << "\tp-value"; } out << "\tlength" << endl; if (method == "pearson") { calcPearson(axes, out); } @@ -349,8 +349,19 @@ int CorrAxesCommand::calcPearson(map >& axes, ofstream& ou double denom = (sqrt(denomTerm1) * sqrt(denomTerm2)); r = numerator / denom; + + if (isnan(r) || isinf(r)) { r = 0.0; } + rValues[k] = r; out << '\t' << r; + + //signifigance calc - http://faculty.vassar.edu/lowry/ch4apx.html + double temp = (1- (r*r)) / (double) (lookupFloat.size()-2); + temp = sqrt(temp); + double sig = r / temp; + if (isnan(sig) || isinf(sig)) { sig = 0.0; } + + out << '\t' << sig; } double sum = 0; @@ -515,9 +526,20 @@ int CorrAxesCommand::calcSpearman(map >& axes, ofstream& o p = (SX2 + SY2 - di) / (2.0 * sqrt((SX2*SY2))); + if (isnan(p) || isinf(p)) { p = 0.0; } + out << '\t' << p; pValues[j] = p; + + //signifigance calc - http://en.wikipedia.org/wiki/Spearman%27s_rank_correlation_coefficient + double temp = (lookupFloat.size()-2) / (double) (1- (p*p)); + temp = sqrt(temp); + double sig = p*temp; + if (isnan(sig) || isinf(sig)) { sig = 0.0; } + + out << '\t' << sig; + } double sum = 0; @@ -651,10 +673,21 @@ int CorrAxesCommand::calcKendall(map >& axes, ofstream& ou } double p = (numCoor - numDisCoor) / (float) count; - + if (isnan(p) || isinf(p)) { p = 0.0; } + out << '\t' << p; pValues[j] = p; - + + //calc signif - zA - http://en.wikipedia.org/wiki/Kendall_tau_rank_correlation_coefficient#Significance_tests + double numer = 3.0 * (numCoor - numDisCoor); + int n = scores[j].size(); + double denom = n * (n-1) * (2*n + 5) / (double) 2.0; + denom = sqrt(denom); + double sig = numer / denom; + + if (isnan(sig) || isinf(sig)) { sig = 0.0; } + + out << '\t' << sig; } double sum = 0; diff --git a/getgroupscommand.cpp b/getgroupscommand.cpp index 432894b..5ed31dd 100644 --- a/getgroupscommand.cpp +++ b/getgroupscommand.cpp @@ -20,6 +20,7 @@ vector GetGroupsCommand::setParameters(){ CommandParameter pshared("shared", "InputTypes", "", "", "none", "sharedGroup", "none",false,false); parameters.push_back(pshared); CommandParameter pname("name", "InputTypes", "", "", "none", "none", "none",false,false); parameters.push_back(pname); CommandParameter pgroup("group", "InputTypes", "", "", "none", "sharedGroup", "FNGLT",false,false); parameters.push_back(pgroup); + CommandParameter pdesign("design", "InputTypes", "", "", "none", "sharedGroup", "FNGLT",false,false); parameters.push_back(pdesign); CommandParameter plist("list", "InputTypes", "", "", "none", "none", "FNGLT",false,false); parameters.push_back(plist); CommandParameter ptaxonomy("taxonomy", "InputTypes", "", "", "none", "none", "FNGLT",false,false); parameters.push_back(ptaxonomy); CommandParameter paccnos("accnos", "InputTypes", "", "", "none", "none", "none",false,false); parameters.push_back(paccnos); @@ -40,9 +41,9 @@ vector GetGroupsCommand::setParameters(){ string GetGroupsCommand::getHelpString(){ try { string helpString = ""; - helpString += "The get.groups command selects sequences from a specfic group or set of groups from the following file types: fasta, name, group, list, taxonomy or shared file.\n"; + helpString += "The get.groups command selects sequences from a specfic group or set of groups from the following file types: fasta, name, group, list, taxonomy, design or shared file.\n"; helpString += "It outputs a file containing the sequences in the those specified groups, or a sharedfile containing only those groups.\n"; - helpString += "The get.groups command parameters are accnos, fasta, name, group, list, taxonomy, shared and groups. The group parameter is required, unless you have a current group file, or are using a shared file.\n"; + helpString += "The get.groups command parameters are accnos, fasta, name, group, list, taxonomy, shared, design and groups. The group parameter is required, unless you have a current group file, or are using a shared file.\n"; helpString += "You must also provide an accnos containing the list of groups to get or set the groups parameter to the groups you wish to select.\n"; helpString += "The groups parameter allows you to specify which of the groups in your groupfile you would like. You can separate group names with dashes.\n"; helpString += "The get.groups command should be in the following format: get.groups(accnos=yourAccnos, fasta=yourFasta, group=yourGroupFile).\n"; @@ -68,6 +69,7 @@ GetGroupsCommand::GetGroupsCommand(){ outputTypes["group"] = tempOutNames; outputTypes["list"] = tempOutNames; outputTypes["shared"] = tempOutNames; + outputTypes["design"] = tempOutNames; } catch(exception& e) { m->errorOut(e, "GetGroupsCommand", "GetGroupsCommand"); @@ -105,6 +107,7 @@ GetGroupsCommand::GetGroupsCommand(string option) { outputTypes["group"] = tempOutNames; outputTypes["list"] = tempOutNames; outputTypes["shared"] = tempOutNames; + outputTypes["design"] = tempOutNames; //if the user changes the output directory command factory will send this info to us in the output parameter @@ -170,6 +173,14 @@ GetGroupsCommand::GetGroupsCommand(string option) { //if the user has not given a path then, add inputdir. else leave path alone. if (path == "") { parameters["shared"] = inputDir + it->second; } } + + it = parameters.find("design"); + //user has given a template file + if(it != parameters.end()){ + path = m->hasPath(it->second); + //if the user has not given a path then, add inputdir. else leave path alone. + if (path == "") { parameters["design"] = inputDir + it->second; } + } } @@ -219,9 +230,14 @@ GetGroupsCommand::GetGroupsCommand(string option) { groupfile = validParameter.validFile(parameters, "group", true); if (groupfile == "not open") { groupfile = ""; abort = true; } else if (groupfile == "not found") { groupfile = ""; } - else { m->setGroupFile(groupfile); } - - if ((sharedfile == "") && (groupfile == "")) { + else { m->setGroupFile(groupfile); } + + designfile = validParameter.validFile(parameters, "design", true); + if (designfile == "not open") { designfile = ""; abort = true; } + else if (designfile == "not found") { designfile = ""; } + else { m->setDesignFile(designfile); } + + if ((sharedfile == "") && (groupfile == "") && (designfile == "")) { //is there are current file available for any of these? if ((namefile != "") || (fastafile != "") || (listfile != "") || (taxfile != "")) { //give priority to group, then shared @@ -242,7 +258,11 @@ GetGroupsCommand::GetGroupsCommand(string option) { groupfile = m->getGroupFile(); if (groupfile != "") { m->mothurOut("Using " + groupfile + " as input file for the group parameter."); m->mothurOutEndLine(); } else { - m->mothurOut("You have no current groupfile or sharedfile and one is required."); m->mothurOutEndLine(); abort = true; + designfile = m->getDesignFile(); + if (designfile != "") { m->mothurOut("Using " + designfile + " as input file for the design parameter."); m->mothurOutEndLine(); } + else { + m->mothurOut("You have no current groupfile or sharedfile or designfile and one is required."); m->mothurOutEndLine(); abort = true; + } } } } @@ -250,7 +270,7 @@ GetGroupsCommand::GetGroupsCommand(string option) { if ((accnosfile == "") && (Groups.size() == 0)) { m->mothurOut("You must provide an accnos file or specify groups using the groups parameter."); m->mothurOutEndLine(); abort = true; } - if ((fastafile == "") && (namefile == "") && (groupfile == "") && (sharedfile == "") && (listfile == "") && (taxfile == "")) { m->mothurOut("You must provide at least one of the following: fasta, name, taxonomy, group, shared or list."); m->mothurOutEndLine(); abort = true; } + if ((fastafile == "") && (namefile == "") && (groupfile == "") && (designfile == "") && (sharedfile == "") && (listfile == "") && (taxfile == "")) { m->mothurOut("You must provide at least one of the following: fasta, name, taxonomy, group, shared, design or list."); m->mothurOutEndLine(); abort = true; } if ((groupfile == "") && ((namefile != "") || (fastafile != "") || (listfile != "") || (taxfile != ""))) { m->mothurOut("If using a fasta, name, taxonomy, group or list, then you must provide a group file."); m->mothurOutEndLine(); abort = true; } if ((namefile == "") && ((fastafile != "") || (taxfile != ""))){ @@ -302,6 +322,7 @@ int GetGroupsCommand::execute(){ if (listfile != "") { readList(); } if (taxfile != "") { readTax(); } if (sharedfile != "") { readShared(); } + if (designfile != "") { readDesign(); } if (m->control_pressed) { for (int i = 0; i < outputNames.size(); i++) { m->mothurRemove(outputNames[i]); } return 0; } @@ -343,6 +364,11 @@ int GetGroupsCommand::execute(){ if (itTypes != outputTypes.end()) { if ((itTypes->second).size() != 0) { current = (itTypes->second)[0]; m->setSharedFile(current); } } + + itTypes = outputTypes.find("design"); + if (itTypes != outputTypes.end()) { + if ((itTypes->second).size() != 0) { current = (itTypes->second)[0]; m->setDesignFile(current); } + } } return 0; @@ -688,6 +714,56 @@ int GetGroupsCommand::readGroup(){ exit(1); } } +//********************************************************************************************************************** +int GetGroupsCommand::readDesign(){ + try { + string thisOutputDir = outputDir; + if (outputDir == "") { thisOutputDir += m->hasPath(designfile); } + string outputFileName = thisOutputDir + m->getRootName(m->getSimpleName(designfile)) + "pick" + m->getExtension(designfile); + + ofstream out; + m->openOutputFile(outputFileName, out); + + ifstream in; + m->openInputFile(designfile, in); + string name, group; + + bool wroteSomething = false; + int selectedCount = 0; + + while(!in.eof()){ + if (m->control_pressed) { in.close(); out.close(); m->mothurRemove(outputFileName); return 0; } + + in >> name; //read from first column + in >> group; //read from second column + + //if this name is in the accnos file + if (m->inUsersGroups(name, Groups)) { + wroteSomething = true; + out << name << '\t' << group << endl; + selectedCount++; + } + + m->gobble(in); + } + in.close(); + out.close(); + + if (wroteSomething == false) { m->mothurOut("Your file does NOT contain groups from the groups you wish to get."); m->mothurOutEndLine(); } + outputTypes["design"].push_back(outputFileName); outputNames.push_back(outputFileName); + + m->mothurOut("Selected " + toString(selectedCount) + " groups from your design file."); m->mothurOutEndLine(); + + + return 0; + } + catch(exception& e) { + m->errorOut(e, "GetGroupsCommand", "readDesign"); + exit(1); + } +} + + //********************************************************************************************************************** int GetGroupsCommand::readTax(){ try { diff --git a/getgroupscommand.h b/getgroupscommand.h index e0fbad8..224f4f0 100644 --- a/getgroupscommand.h +++ b/getgroupscommand.h @@ -27,7 +27,7 @@ public: string getCommandCategory() { return "OTU-Based Approaches"; } string getHelpString(); string getCitation() { return "http://www.mothur.org/wiki/Get.groups"; } - string getDescription() { return "gets sequences from a list, fasta, name, group or taxonomy file from a given group or set of groups"; } + string getDescription() { return "gets sequences from a list, fasta, name, group, shared, design or taxonomy file from a given group or set of groups"; } int execute(); @@ -39,7 +39,7 @@ private: map uniqueToRedundant; //if a namefile is given and the first column name is not selected //then the other files need to change the unique name in their file to match. //only add the names that need to be changed to keep the map search quick - string accnosfile, fastafile, namefile, groupfile, listfile, taxfile, outputDir, groups, sharedfile; + string accnosfile, fastafile, namefile, groupfile, listfile, designfile, taxfile, outputDir, groups, sharedfile; bool abort; vector outputNames, Groups; GroupMap* groupMap; @@ -52,6 +52,7 @@ private: int readTax(); int fillNames(); int readShared(); + int readDesign(); }; diff --git a/removegroupscommand.cpp b/removegroupscommand.cpp index f35da2e..49674f5 100644 --- a/removegroupscommand.cpp +++ b/removegroupscommand.cpp @@ -20,6 +20,7 @@ vector RemoveGroupsCommand::setParameters(){ CommandParameter pshared("shared", "InputTypes", "", "", "none", "sharedGroup", "none",false,false); parameters.push_back(pshared); CommandParameter pname("name", "InputTypes", "", "", "none", "none", "none",false,false); parameters.push_back(pname); CommandParameter pgroup("group", "InputTypes", "", "", "none", "sharedGroup", "FNGLT",false,false); parameters.push_back(pgroup); + CommandParameter pdesign("design", "InputTypes", "", "", "none", "sharedGroup", "FNGLT",false,false); parameters.push_back(pdesign); CommandParameter plist("list", "InputTypes", "", "", "none", "none", "FNGLT",false,false); parameters.push_back(plist); CommandParameter ptaxonomy("taxonomy", "InputTypes", "", "", "none", "none", "FNGLT",false,false); parameters.push_back(ptaxonomy); CommandParameter paccnos("accnos", "InputTypes", "", "", "none", "none", "none",false,false); parameters.push_back(paccnos); @@ -40,9 +41,9 @@ vector RemoveGroupsCommand::setParameters(){ string RemoveGroupsCommand::getHelpString(){ try { string helpString = ""; - helpString += "The remove.groups command removes sequences from a specfic group or set of groups from the following file types: fasta, name, group, list, taxonomy or sharedfile.\n"; + helpString += "The remove.groups command removes sequences from a specfic group or set of groups from the following file types: fasta, name, group, list, taxonomy, design or sharedfile.\n"; helpString += "It outputs a file containing the sequences NOT in the those specified groups, or with a sharedfile eliminates the groups you selected.\n"; - helpString += "The remove.groups command parameters are accnos, fasta, name, group, list, taxonomy, shared and groups. The group parameter is required, unless you have a current group file or are using a sharedfile.\n"; + helpString += "The remove.groups command parameters are accnos, fasta, name, group, list, taxonomy, shared, design and groups. The group parameter is required, unless you have a current group file or are using a sharedfile.\n"; helpString += "You must also provide an accnos containing the list of groups to remove or set the groups parameter to the groups you wish to remove.\n"; helpString += "The groups parameter allows you to specify which of the groups in your groupfile you would like removed. You can separate group names with dashes.\n"; helpString += "The remove.groups command should be in the following format: remove.groups(accnos=yourAccnos, fasta=yourFasta, group=yourGroupFile).\n"; @@ -69,6 +70,7 @@ RemoveGroupsCommand::RemoveGroupsCommand(){ outputTypes["group"] = tempOutNames; outputTypes["list"] = tempOutNames; outputTypes["shared"] = tempOutNames; + outputTypes["design"] = tempOutNames; } catch(exception& e) { m->errorOut(e, "RemoveGroupsCommand", "RemoveGroupsCommand"); @@ -106,6 +108,7 @@ RemoveGroupsCommand::RemoveGroupsCommand(string option) { outputTypes["group"] = tempOutNames; outputTypes["list"] = tempOutNames; outputTypes["shared"] = tempOutNames; + outputTypes["design"] = tempOutNames; //if the user changes the output directory command factory will send this info to us in the output parameter @@ -171,6 +174,14 @@ RemoveGroupsCommand::RemoveGroupsCommand(string option) { //if the user has not given a path then, add inputdir. else leave path alone. if (path == "") { parameters["shared"] = inputDir + it->second; } } + + it = parameters.find("design"); + //user has given a template file + if(it != parameters.end()){ + path = m->hasPath(it->second); + //if the user has not given a path then, add inputdir. else leave path alone. + if (path == "") { parameters["design"] = inputDir + it->second; } + } } @@ -204,6 +215,11 @@ RemoveGroupsCommand::RemoveGroupsCommand(string option) { if (taxfile == "not open") { taxfile = ""; abort = true; } else if (taxfile == "not found") { taxfile = ""; } else { m->setTaxonomyFile(taxfile); } + + designfile = validParameter.validFile(parameters, "design", true); + if (designfile == "not open") { designfile = ""; abort = true; } + else if (designfile == "not found") { designfile = ""; } + else { m->setDesignFile(designfile); } groups = validParameter.validFile(parameters, "groups", false); if (groups == "not found") { groups = ""; } @@ -222,7 +238,7 @@ RemoveGroupsCommand::RemoveGroupsCommand(string option) { else if (groupfile == "not found") { groupfile = ""; } else { m->setGroupFile(groupfile); } - if ((sharedfile == "") && (groupfile == "")) { + if ((sharedfile == "") && (groupfile == "") && (designfile == "")) { //is there are current file available for any of these? if ((namefile != "") || (fastafile != "") || (listfile != "") || (taxfile != "")) { //give priority to group, then shared @@ -243,7 +259,11 @@ RemoveGroupsCommand::RemoveGroupsCommand(string option) { groupfile = m->getGroupFile(); if (groupfile != "") { m->mothurOut("Using " + groupfile + " as input file for the group parameter."); m->mothurOutEndLine(); } else { - m->mothurOut("You have no current groupfile or sharedfile and one is required."); m->mothurOutEndLine(); abort = true; + designfile = m->getDesignFile(); + if (designfile != "") { m->mothurOut("Using " + designfile + " as input file for the design parameter."); m->mothurOutEndLine(); } + else { + m->mothurOut("You have no current groupfile or sharedfile or designfile and one is required."); m->mothurOutEndLine(); abort = true; + } } } } @@ -251,7 +271,7 @@ RemoveGroupsCommand::RemoveGroupsCommand(string option) { if ((accnosfile == "") && (Groups.size() == 0)) { m->mothurOut("You must provide an accnos file containing group names or specify groups using the groups parameter."); m->mothurOutEndLine(); abort = true; } - if ((fastafile == "") && (namefile == "") && (groupfile == "") && (sharedfile == "") && (listfile == "") && (taxfile == "")) { m->mothurOut("You must provide at least one of the following: fasta, name, taxonomy, group, shared or list."); m->mothurOutEndLine(); abort = true; } + if ((fastafile == "") && (namefile == "") && (groupfile == "") && (sharedfile == "") && (designfile == "") && (listfile == "") && (taxfile == "")) { m->mothurOut("You must provide at least one of the following: fasta, name, taxonomy, group, shared, design or list."); m->mothurOutEndLine(); abort = true; } if ((groupfile == "") && ((namefile != "") || (fastafile != "") || (listfile != "") || (taxfile != ""))) { m->mothurOut("If using a fasta, name, taxonomy, group or list, then you must provide a group file."); m->mothurOutEndLine(); abort = true; } if ((namefile == "") && ((fastafile != "") || (taxfile != ""))){ @@ -303,6 +323,7 @@ int RemoveGroupsCommand::execute(){ if (listfile != "") { readList(); } if (taxfile != "") { readTax(); } if (sharedfile != "") { readShared(); } + if (designfile != "") { readDesign(); } if (m->control_pressed) { for (int i = 0; i < outputNames.size(); i++) { m->mothurRemove(outputNames[i]); } return 0; } @@ -343,6 +364,11 @@ int RemoveGroupsCommand::execute(){ if (itTypes != outputTypes.end()) { if ((itTypes->second).size() != 0) { current = (itTypes->second)[0]; m->setSharedFile(current); } } + + itTypes = outputTypes.find("design"); + if (itTypes != outputTypes.end()) { + if ((itTypes->second).size() != 0) { current = (itTypes->second)[0]; m->setDesignFile(current); } + } } return 0; @@ -713,6 +739,54 @@ int RemoveGroupsCommand::readGroup(){ exit(1); } } +//********************************************************************************************************************** +int RemoveGroupsCommand::readDesign(){ + try { + string thisOutputDir = outputDir; + if (outputDir == "") { thisOutputDir += m->hasPath(designfile); } + string outputFileName = thisOutputDir + m->getRootName(m->getSimpleName(designfile)) + "pick" + m->getExtension(designfile); + + ofstream out; + m->openOutputFile(outputFileName, out); + + ifstream in; + m->openInputFile(designfile, in); + string name, group; + + bool wroteSomething = false; + int removedCount = 0; + + while(!in.eof()){ + if (m->control_pressed) { in.close(); out.close(); m->mothurRemove(outputFileName); return 0; } + + in >> name; //read from first column + in >> group; //read from second column + + //if this name is in the accnos file + if (!(m->inUsersGroups(name, Groups))) { + wroteSomething = true; + out << name << '\t' << group << endl; + }else { removedCount++; } + + m->gobble(in); + } + in.close(); + out.close(); + + if (wroteSomething == false) { m->mothurOut("Your file contains only groups from the groups you wish to remove."); m->mothurOutEndLine(); } + outputTypes["design"].push_back(outputFileName); outputNames.push_back(outputFileName); + + m->mothurOut("Removed " + toString(removedCount) + " groups from your design file."); m->mothurOutEndLine(); + + + return 0; + } + catch(exception& e) { + m->errorOut(e, "RemoveGroupsCommand", "readDesign"); + exit(1); + } +} + //********************************************************************************************************************** int RemoveGroupsCommand::readTax(){ try { diff --git a/removegroupscommand.h b/removegroupscommand.h index 103ab66..24062ba 100644 --- a/removegroupscommand.h +++ b/removegroupscommand.h @@ -27,7 +27,7 @@ public: string getCommandCategory() { return "OTU-Based Approaches"; } string getHelpString(); string getCitation() { return "http://www.mothur.org/wiki/Remove.groups"; } - string getDescription() { return "removes sequences from a list, fasta, name, group or taxonomy file from a given group or set of groups"; } + string getDescription() { return "removes sequences from a list, fasta, name, group, shared, design or taxonomy file from a given group or set of groups"; } int execute(); void help() { m->mothurOut(getHelpString()); } @@ -35,7 +35,7 @@ public: private: set names; - string accnosfile, fastafile, namefile, groupfile, listfile, taxfile, outputDir, groups, sharedfile; + string accnosfile, fastafile, namefile, groupfile, designfile, listfile, taxfile, outputDir, groups, sharedfile; bool abort; vector outputNames, Groups; GroupMap* groupMap; @@ -52,6 +52,7 @@ private: int readList(); int readTax(); int fillNames(); + int readDesign(); }; -- 2.39.2