//
// printPedigreeTableAsTabDelimited: Prints core and optional non-core columns:
//
void DataTable::printPedigreeTableAsTabDelimited(std::string filename) const{
// Get the Core Column Indices:
unsigned familyColumnIndex = getColumnOrdinal("FamilyId");
unsigned individualColumnIndex = getColumnOrdinal("IndividualId");
unsigned genderColumnIndex = getColumnOrdinal("Gender");
unsigned fatherColumnIndex = getColumnOrdinal("Father");
unsigned motherColumnIndex = getColumnOrdinal("Mother");
int firstNameColumnIndex = (columnExists("FirstName")?(int)getColumnOrdinal("FirstName"):COLUMN_IS_MISSING);
int lastNameColumnIndex = (columnExists("LastName")?(int)getColumnOrdinal("LastName"):COLUMN_IS_MISSING);
unsigned j;
std::ofstream pedfile(filename.c_str());
if(pedfile.is_open()){
// Print the header row:
pedfile << "FamilyId\t";
pedfile << "IndividualId\t";
pedfile << "Gender\t";
pedfile << "Father\t";
pedfile << "Mother\t";
pedfile << "Deceased\t";
pedfile << "Proband\t";
pedfile << "DOB\t";
pedfile << "MZTwin\t";
pedfile << "DZTwin\t";
pedfile << "Affected\t";
pedfile << "FirstName\t";
pedfile << "LastName\n";
// Print the data by rows
for(j=0;j<_rows;j++){
pedfile << (*_columnVector[familyColumnIndex])->get(j) << "\t";
pedfile << (*_columnVector[individualColumnIndex])->get(j) << "\t";
pedfile << (*_columnVector[genderColumnIndex])->get(j) << "\t";
pedfile << (*_columnVector[fatherColumnIndex])->get(j) << "\t";
pedfile << (*_columnVector[motherColumnIndex])->get(j) << "\t";
if(_deceasedColumnIndex != COLUMN_IS_MISSING) pedfile << (*_columnVector[_deceasedColumnIndex])->get(j) << "\t";
else pedfile << "." << "\t";
if(_probandColumnIndex != COLUMN_IS_MISSING) pedfile << (*_columnVector[_probandColumnIndex])->get(j) << "\t";
else pedfile << "." << "\t";
if(_dobColumnIndex != COLUMN_IS_MISSING) pedfile << (*_columnVector[_dobColumnIndex])->get(j) << "\t";
else pedfile << "." << "\t";
if(_mzTwinColumnIndex != COLUMN_IS_MISSING) pedfile << (*_columnVector[_mzTwinColumnIndex])->get(j) << "\t";
else pedfile << "." << "\t";
if(_dzTwinColumnIndex != COLUMN_IS_MISSING) pedfile << (*_columnVector[_dzTwinColumnIndex])->get(j) << "\t";
else pedfile << "." << "\t";
if(_affectedColumnIndex != COLUMN_IS_MISSING) pedfile << (*_columnVector[_affectedColumnIndex])->get(j) << "\t";
else pedfile << "." << "\t";
if(firstNameColumnIndex != COLUMN_IS_MISSING) pedfile << (*_columnVector[firstNameColumnIndex])->get(j) << "\t";
else pedfile << "." << "\t";
if(lastNameColumnIndex != COLUMN_IS_MISSING) pedfile << (*_columnVector[lastNameColumnIndex])->get(j) << "\n";
else pedfile << "." << "\n";
}
pedfile.close();
}
}
void RPedigree:: output(char* ped){
// Authors: Rohan L. Fernando
// (2005)
// Contributors:
ofstream pedfile(ped);
SafeSTLVector<PNode*>::iterator vecit;
pedfile.setf(ios::fixed | ios::right);
for (vecit=pedVector.begin();vecit!=pedVector.end();vecit++){
pedfile << setw(10) << (*vecit)->ind
<< setw(10) << (*vecit)->sire
<< setw(10) << (*vecit)->dam
<< setw(20) << ((*vecit)->ind_str).c_str()
<< setw(20) << (*vecit)->f << endl;
}
}
int main(int argc, char * argv[])
{
printf("PedWipe - (c) 2000 Goncalo Abecasis\n"
"Automatically wipe out genotypes from a pedigree file\n\n");
String pedfile("merlin.ped");
String datafile("merlin.dat");
String errorfile("merlin.err");
bool showTallies = false;
ParameterList pl;
pl.Add(new StringParameter('d', "Data File", datafile));
pl.Add(new StringParameter('p', "Pedigree File", pedfile));
pl.Add(new StringParameter('e', "Errors File", errorfile));
pl.Add(new SwitchParameter('t', "Show Tallies", showTallies));
pl.Read(argc, argv);
pl.Status();
Pedigree ped;
ped.Prepare(datafile);
ped.Load(pedfile);
StringArray errors, tokens;
errors.Read(errorfile);
int count = 0;
StringIntMap perMarker, perFamily, perPerson;
for (int i = 1; i < errors.Length(); i++)
{
tokens.Clear();
tokens.AddTokens(errors[i]);
if (tokens.Length() < 3) continue;
Person * person = ped.FindPerson(tokens[0], tokens[1]);
int markerid = ped.LookupMarker(tokens[2]);
if (person == NULL)
{
printf("Person %s.%s not found ... \n",
(const char *) tokens[0], (const char *) tokens[1]);
continue;
}
if (markerid == -1)
{
printf("Marker %s not found ... \n",
(const char *) tokens[2]);
continue;
}
printf("Person %s.%s, marker %s wiped.\n",
(const char *) tokens[0], (const char *) tokens[1],
(const char *) tokens[2]);
person->markers[markerid].one = 0;
person->markers[markerid].two = 0;
perPerson.IncrementCount(tokens[0] + "." + tokens[1]);
perFamily.IncrementCount(tokens[0]);
perMarker.IncrementCount(tokens[2]);
count++;
}
if (perMarker.Length() == 0)
{
printf("No errors found in merlin.err\n");
}
else if (showTallies && count)
{
printf("\nSummary of Errors\n");
printf("=================\n\n");
QuickIndex index;
printf("Per Marker: (average = %.2f)\n"
"-----------------------------\n",
(double) count / (double) ped.markerCount);
index.IndexCounts(perMarker);
index.Reverse();
for (int i = 0; i < perMarker.Length(); i++)
printf(" %3d errors for marker %s\n",
perMarker.GetCount(index[i]),
(const char *) perMarker[index[i]]);
printf("\nPer Family: (average = %.2f)\n"
"----------------------------\n",
(double) count / (double) ped.familyCount);
index.IndexCounts(perFamily);
index.Reverse();
for (int i = 0; i < perFamily.Length(); i++)
printf(" %3d errors for family %s\n",
perFamily.GetCount(index[i]), (const char *) perFamily[index[i]]);
printf("\nPer Person: (average = %.2f)\n"
"----------------------------\n",
(double) count / (double) ped.count);
index.IndexCounts(perPerson);
index.Reverse();
for (int i = 0; i < perPerson.Length(); i++)
printf(" %3d errors for person %s\n",
perPerson.GetCount(index[i]), (const char *) perPerson[index[i]]);
}
printf("\nWriting out edited files [wiped.*] ...\n\n");
if (ped.markerInfoCount)
{
ped.WriteMapFile("wiped.map");
ped.WriteFreqFile("wiped.freq");
}
ped.WriteDataFile("wiped.dat");
ped.WritePedigreeFile("wiped.ped");
}
