X-Git-Url: https://git.donarmstrong.com/?p=mothur.git;a=blobdiff_plain;f=counttable.cpp;h=6eca8ee2a4b5354da54412b9eb6986b6ba244657;hp=bc9d4da05966b6b3fc3ec5fe0e751c21be7f180f;hb=df7e3ff9f68ef157b0328a2d353c3258c5d45d89;hpb=5b72d1cf3fa48730e5bb70d59cced1e43e1fe424 diff --git a/counttable.cpp b/counttable.cpp index bc9d4da..6eca8ee 100644 --- a/counttable.cpp +++ b/counttable.cpp @@ -40,7 +40,13 @@ int CountTable::createTable(set& n, map& g, set& if (itGroup != g.end()) { groupCounts[indexGroupMap[itGroup->second]] = 1; totalGroups[indexGroupMap[itGroup->second]]++; - }else { m->mothurOut("[ERROR]: Your group file does not contain " + seqName + ". Please correct."); m->mothurOutEndLine(); } + }else { + //look for it in names of groups to see if the user accidently used the wrong file + if (m->inUsersGroups(seqName, groups)) { + m->mothurOut("[WARNING]: Your group or design file contains a group named " + seqName + ". Perhaps you are used a group file instead of a design file? A common cause of this is using a tree file that relates your groups (created by the tree.shared command) with a group file that assigns sequences to a group."); m->mothurOutEndLine(); + } + m->mothurOut("[ERROR]: Your group file does not contain " + seqName + ". Please correct."); m->mothurOutEndLine(); + } map::iterator it2 = indexNameMap.find(seqName); if (it2 == indexNameMap.end()) { @@ -52,6 +58,12 @@ int CountTable::createTable(set& n, map& g, set& } } + if (hasGroups) { + for (int i = 0; i < totalGroups.size(); i++) { + if (totalGroups[i] == 0) { m->mothurOut("\nRemoving group: " + groups[i] + " because all sequences have been removed.\n"); removeGroup(groups[i]); i--; } + } + } + return 0; } catch(exception& e) { @@ -125,6 +137,9 @@ int CountTable::createTable(string namefile, string groupfile, bool createGroup) string firstCol, secondCol; in >> firstCol; m->gobble(in); in >> secondCol; m->gobble(in); + m->checkName(firstCol); + m->checkName(secondCol); + vector names; m->splitAtChar(secondCol, names, ','); @@ -179,7 +194,14 @@ int CountTable::createTable(string namefile, string groupfile, bool createGroup) in.close(); if (error) { m->control_pressed = true; } - if (groupfile != "") { delete groupMap; } + else { //check for zero groups + if (hasGroups) { + for (int i = 0; i < totalGroups.size(); i++) { + if (totalGroups[i] == 0) { m->mothurOut("\nRemoving group: " + groups[i] + " because all sequences have been removed.\n"); removeGroup(groups[i]); i--; } + } + } + } + if (groupfile != "") { delete groupMap; } return 0; } @@ -189,7 +211,7 @@ int CountTable::createTable(string namefile, string groupfile, bool createGroup) } } /************************************************************/ -int CountTable::readTable(string file) { +int CountTable::readTable(string file, bool readGroups) { try { filename = file; ifstream in; @@ -205,7 +227,7 @@ int CountTable::readTable(string file) { indexNameMap.clear(); counts.clear(); map originalGroupIndexes; - if (columnHeaders.size() > 2) { hasGroups = true; numGroups = columnHeaders.size() - 2; } + if ((columnHeaders.size() > 2) && readGroups) { hasGroups = true; numGroups = columnHeaders.size() - 2; } for (int i = 2; i < columnHeaders.size(); i++) { groups.push_back(columnHeaders[i]); originalGroupIndexes[i-2] = columnHeaders[i]; totalGroups.push_back(0); } //sort groups to keep consistent with how we store the groups in groupmap sort(groups.begin(), groups.end()); @@ -226,7 +248,13 @@ int CountTable::readTable(string file) { //if group info, then read it vector groupCounts; groupCounts.resize(numGroups, 0); - for (int i = 0; i < numGroups; i++) { int thisIndex = indexGroupMap[originalGroupIndexes[i]]; in >> groupCounts[thisIndex]; m->gobble(in); totalGroups[thisIndex] += groupCounts[thisIndex]; } + if (columnHeaders.size() > 2) { //file contains groups + if (readGroups) { //user wants to save them + for (int i = 0; i < numGroups; i++) { int thisIndex = indexGroupMap[originalGroupIndexes[i]]; in >> groupCounts[thisIndex]; m->gobble(in); totalGroups[thisIndex] += groupCounts[thisIndex]; } + }else { //read and discard + m->getline(in); m->gobble(in); + } + } map::iterator it = indexNameMap.find(name); if (it == indexNameMap.end()) { @@ -243,6 +271,13 @@ int CountTable::readTable(string file) { in.close(); if (error) { m->control_pressed = true; } + else { //check for zero groups + if (hasGroups) { + for (int i = 0; i < totalGroups.size(); i++) { + if (totalGroups[i] == 0) { m->mothurOut("\nRemoving group: " + groups[i] + " because all sequences have been removed.\n"); removeGroup(groups[i]); i--; } + } + } + } return 0; } @@ -260,7 +295,23 @@ int CountTable::printTable(string file) { for (int i = 0; i < groups.size(); i++) { out << groups[i] << '\t'; } out << endl; - for (map::iterator itNames = indexNameMap.begin(); itNames != indexNameMap.end(); itNames++) { + map reverse; //use this to preserve order + for (map::iterator it = indexNameMap.begin(); it !=indexNameMap.end(); it++) { reverse[it->second] = it->first; } + + for (int i = 0; i < totals.size(); i++) { + map::iterator itR = reverse.find(i); + + if (itR != reverse.end()) { //will equal end if seqs were removed because remove just removes from indexNameMap + out << itR->second << '\t' << totals[i] << '\t'; + if (hasGroups) { + for (int j = 0; j < groups.size(); j++) { + out << counts[i][j] << '\t'; + } + } + out << endl; + } + } + /*for (map::iterator itNames = indexNameMap.begin(); itNames != indexNameMap.end(); itNames++) { out << itNames->first << '\t' << totals[itNames->second] << '\t'; if (hasGroups) { @@ -269,7 +320,7 @@ int CountTable::printTable(string file) { } } out << endl; - } + }*/ out.close(); return 0; } @@ -321,6 +372,10 @@ vector CountTable::getGroupCounts(string seqName) { if (hasGroups) { map::iterator it = indexNameMap.find(seqName); if (it == indexNameMap.end()) { + //look for it in names of groups to see if the user accidently used the wrong file + if (m->inUsersGroups(seqName, groups)) { + m->mothurOut("[WARNING]: Your group or design file contains a group named " + seqName + ". Perhaps you are used a group file instead of a design file? A common cause of this is using a tree file that relates your groups (created by the tree.shared command) with a group file that assigns sequences to a group."); m->mothurOutEndLine(); + } m->mothurOut("[ERROR]: " + seqName + " is not in your count table. Please correct.\n"); m->control_pressed = true; }else { temp = counts[it->second]; @@ -341,7 +396,7 @@ int CountTable::getGroupCount(string groupName) { if (hasGroups) { map::iterator it = indexGroupMap.find(groupName); if (it == indexGroupMap.end()) { - m->mothurOut("[ERROR]: " + groupName + " is not in your count table. Please correct.\n"); m->control_pressed = true; + m->mothurOut("[ERROR]: group " + groupName + " is not in your count table. Please correct.\n"); m->control_pressed = true; }else { return totalGroups[it->second]; } @@ -361,11 +416,15 @@ int CountTable::getGroupCount(string seqName, string groupName) { if (hasGroups) { map::iterator it = indexGroupMap.find(groupName); if (it == indexGroupMap.end()) { - m->mothurOut("[ERROR]: " + groupName + " is not in your count table. Please correct.\n"); m->control_pressed = true; + m->mothurOut("[ERROR]: group " + groupName + " is not in your count table. Please correct.\n"); m->control_pressed = true; }else { map::iterator it2 = indexNameMap.find(seqName); if (it2 == indexNameMap.end()) { - m->mothurOut("[ERROR]: " + seqName + " is not in your count table. Please correct.\n"); m->control_pressed = true; + //look for it in names of groups to see if the user accidently used the wrong file + if (m->inUsersGroups(seqName, groups)) { + m->mothurOut("[WARNING]: Your group or design file contains a group named " + seqName + ". Perhaps you are used a group file instead of a design file? A common cause of this is using a tree file that relates your groups (created by the tree.shared command) with a group file that assigns sequences to a group."); m->mothurOutEndLine(); + } + m->mothurOut("[ERROR]: seq " + seqName + " is not in your count table. Please correct.\n"); m->control_pressed = true; }else { return counts[it2->second][it->second]; } @@ -390,6 +449,10 @@ int CountTable::setAbund(string seqName, string groupName, int num) { }else { map::iterator it2 = indexNameMap.find(seqName); if (it2 == indexNameMap.end()) { + //look for it in names of groups to see if the user accidently used the wrong file + if (m->inUsersGroups(seqName, groups)) { + m->mothurOut("[WARNING]: Your group or design file contains a group named " + seqName + ". Perhaps you are used a group file instead of a design file? A common cause of this is using a tree file that relates your groups (created by the tree.shared command) with a group file that assigns sequences to a group."); m->mothurOutEndLine(); + } m->mothurOut("[ERROR]: " + seqName + " is not in your count table. Please correct.\n"); m->control_pressed = true; }else { int oldCount = counts[it2->second][it->second]; @@ -457,6 +520,61 @@ int CountTable::addGroup(string groupName) { } } /************************************************************/ +//remove group +int CountTable::removeGroup(string groupName) { + try { + if (hasGroups) { + //save for later in case removing a group means we need to remove a seq. + map reverse; + for (map::iterator it = indexNameMap.begin(); it !=indexNameMap.end(); it++) { reverse[it->second] = it->first; } + + map::iterator it = indexGroupMap.find(groupName); + if (it == indexGroupMap.end()) { + m->mothurOut("[ERROR]: " + groupName + " is not in your count table. Please correct.\n"); m->control_pressed = true; + }else { + int indexOfGroupToRemove = it->second; + map currentGroupIndex = indexGroupMap; + vector newGroups; + for (int i = 0; i < groups.size(); i++) { + if (groups[i] != groupName) { + newGroups.push_back(groups[i]); + indexGroupMap[groups[i]] = newGroups.size()-1; + } + } + indexGroupMap.erase(groupName); + groups = newGroups; + totalGroups.erase(totalGroups.begin()+indexOfGroupToRemove); + + int thisIndex = 0; + map newIndexNameMap; + for (int i = 0; i < counts.size(); i++) { + int num = counts[i][indexOfGroupToRemove]; + counts[i].erase(counts[i].begin()+indexOfGroupToRemove); + totals[i] -= num; + total -= num; + if (totals[i] == 0) { //your sequences are only from the group we want to remove, then remove you. + counts.erase(counts.begin()+i); + totals.erase(totals.begin()+i); + uniques--; + i--; + } + newIndexNameMap[reverse[thisIndex]] = i; + thisIndex++; + } + indexNameMap = newIndexNameMap; + + if (groups.size() == 0) { hasGroups = false; } + } + }else { m->mothurOut("[ERROR]: your count table does not contain group information, can not remove group " + groupName + ".\n"); m->control_pressed = true; } + + return 0; + } + catch(exception& e) { + m->errorOut(e, "CountTable", "removeGroup"); + exit(1); + } +} +/************************************************************/ //vector of groups for the seq vector CountTable::getGroups(string seqName) { try { @@ -482,6 +600,12 @@ int CountTable::renameSeq(string oldSeqName, string newSeqName) { map::iterator it = indexNameMap.find(oldSeqName); if (it == indexNameMap.end()) { + if (hasGroupInfo()) { + //look for it in names of groups to see if the user accidently used the wrong file + if (m->inUsersGroups(oldSeqName, groups)) { + m->mothurOut("[WARNING]: Your group or design file contains a group named " + oldSeqName + ". Perhaps you are used a group file instead of a design file? A common cause of this is using a tree file that relates your groups (created by the tree.shared command) with a group file that assigns sequences to a group."); m->mothurOutEndLine(); + } + } m->mothurOut("[ERROR]: " + oldSeqName + " is not in your count table. Please correct.\n"); m->control_pressed = true; }else { int index = it->second; @@ -504,6 +628,12 @@ int CountTable::getNumSeqs(string seqName) { map::iterator it = indexNameMap.find(seqName); if (it == indexNameMap.end()) { + if (hasGroupInfo()) { + //look for it in names of groups to see if the user accidently used the wrong file + if (m->inUsersGroups(seqName, groups)) { + m->mothurOut("[WARNING]: Your group or design file contains a group named " + seqName + ". Perhaps you are used a group file instead of a design file? A common cause of this is using a tree file that relates your groups (created by the tree.shared command) with a group file that assigns sequences to a group."); m->mothurOutEndLine(); + } + } m->mothurOut("[ERROR]: " + seqName + " is not in your count table. Please correct.\n"); m->control_pressed = true; }else { return totals[it->second]; @@ -523,6 +653,12 @@ int CountTable::get(string seqName) { map::iterator it = indexNameMap.find(seqName); if (it == indexNameMap.end()) { + if (hasGroupInfo()) { + //look for it in names of groups to see if the user accidently used the wrong file + if (m->inUsersGroups(seqName, groups)) { + m->mothurOut("[WARNING]: Your group or design file contains a group named " + seqName + ". Perhaps you are used a group file instead of a design file? A common cause of this is using a tree file that relates your groups (created by the tree.shared command) with a group file that assigns sequences to a group."); m->mothurOutEndLine(); + } + } m->mothurOut("[ERROR]: " + seqName + " is not in your count table. Please correct.\n"); m->control_pressed = true; }else { return it->second; } @@ -560,7 +696,7 @@ int CountTable::push_back(string seqName) { int CountTable::remove(string seqName) { try { map::iterator it = indexNameMap.find(seqName); - if (it == indexNameMap.end()) { + if (it != indexNameMap.end()) { uniques--; if (hasGroups){ //remove this sequences counts from group totals for (int i = 0; i < totalGroups.size(); i++) { totalGroups[i] -= counts[it->second][i]; counts[it->second][i] = 0; } @@ -569,6 +705,12 @@ int CountTable::remove(string seqName) { total -= thisTotal; indexNameMap.erase(it); }else { + if (hasGroupInfo()) { + //look for it in names of groups to see if the user accidently used the wrong file + if (m->inUsersGroups(seqName, groups)) { + m->mothurOut("[WARNING]: Your group or design file contains a group named " + seqName + ". Perhaps you are used a group file instead of a design file? A common cause of this is using a tree file that relates your groups (created by the tree.shared command) with a group file that assigns sequences to a group."); m->mothurOutEndLine(); + } + } m->mothurOut("[ERROR]: Your count table contains does not include " + seqName + ", cannot remove."); m->mothurOutEndLine(); m->control_pressed = true; } @@ -661,6 +803,22 @@ vector CountTable::getNamesOfSeqs() { } } /************************************************************/ +//returns the names of all unique sequences in file mapped to their seqCounts +map CountTable::getNameMap() { + try { + map names; + for (map::iterator it = indexNameMap.begin(); it != indexNameMap.end(); it++) { + names[it->first] = totals[it->second]; + } + + return names; + } + catch(exception& e) { + m->errorOut(e, "CountTable", "getNameMap"); + exit(1); + } +} +/************************************************************/ //returns the names of all unique sequences in file vector CountTable::getNamesOfSeqs(string group) { try { @@ -689,10 +847,22 @@ int CountTable::mergeCounts(string seq1, string seq2) { try { map::iterator it = indexNameMap.find(seq1); if (it == indexNameMap.end()) { + if (hasGroupInfo()) { + //look for it in names of groups to see if the user accidently used the wrong file + if (m->inUsersGroups(seq1, groups)) { + m->mothurOut("[WARNING]: Your group or design file contains a group named " + seq1 + ". Perhaps you are used a group file instead of a design file? A common cause of this is using a tree file that relates your groups (created by the tree.shared command) with a group file that assigns sequences to a group."); m->mothurOutEndLine(); + } + } m->mothurOut("[ERROR]: " + seq1 + " is not in your count table. Please correct.\n"); m->control_pressed = true; }else { map::iterator it2 = indexNameMap.find(seq2); if (it2 == indexNameMap.end()) { + if (hasGroupInfo()) { + //look for it in names of groups to see if the user accidently used the wrong file + if (m->inUsersGroups(seq2, groups)) { + m->mothurOut("[WARNING]: Your group or design file contains a group named " + seq2 + ". Perhaps you are used a group file instead of a design file? A common cause of this is using a tree file that relates your groups (created by the tree.shared command) with a group file that assigns sequences to a group."); m->mothurOutEndLine(); + } + } m->mothurOut("[ERROR]: " + seq2 + " is not in your count table. Please correct.\n"); m->control_pressed = true; }else { //merge data