Rcpp::StringMatrix proc_feature( Rcpp::StringVector myBed,
// std::vector< int > POS,
Rcpp::StringMatrix myData,
int fill_missing
){
// myBed is a StringVector:
// myBed(0) = chrom
// myBed(1) = chromStart
// myBed(2) = chromEnd
// myBed(3) = name
// myData is a StringMatrix
// column 0 = CHROM
// column 1 = POS
// Convert POS to ints
std::vector< int > POS = get_pos(myData);
// Create an empty matrix to return in exceptions.
Rcpp::StringMatrix MT_matrix(0, myData.ncol());
// Rcpp::StringVector myBed includes start and stop integer coordinates.
// Convert Rcpp::StringVector elements to int
std::string temp = Rcpp::as< std::string >( myBed(1) );
int start = atoi(temp.c_str());
temp = Rcpp::as< std::string >( myBed(2) );
int end = atoi(temp.c_str());
// Manage if feature is on reverse strand
if(end < start){
int tmp = start;
start = end;
end = tmp;
}
// Increment i so that chromosome in myData
// matches the chromosome in the single BED record.
int i = 0; // Data row counter
while ( myData(i,0) != myBed(0) && i < myData.nrow() ){
i++;
}
// If we didn't find the chromosome return an empty matrix.
if( i == myData.nrow() & fill_missing != 1 ){
return MT_matrix;
}
// Rcpp::Rcout << " Found CHROM " << myBed(0) << " in myData CHROM:" << myData(i,0) << " POS:" << myData(i,1) << "\n";
// We should now have i at the correct chromosome in myData.
// POS is the integer recast of POS in myData.
// We can now increment to the correct position in the chromosome
// by incrementing to the start of teh annotation.
//
// Increment to POS.
while( myData(i,0) == myBed(0) && POS[i] < start && i < myData.nrow() ){
i++;
}
// If we didn't find the POS return an empty matrix.
if( i == myData.nrow() & fill_missing != 1 ){
return MT_matrix;
}
// Rcpp::Rcout << " Found CHROM " << myBed(0) << " in myData POS:" << POS[i] << "\n";
// Increment to the end of the feature
int j=i;
while( myData(j,0) == myBed(0) && POS[j] <= end && j < myData.nrow() ){
j++;
}
// Rcpp::Rcout << " Found end of feature at: " << POS[j-1] << "\n";
// We now have the information to declare a return matrix
// and populate it.
if( fill_missing != 1 ){
// Do not fill missubg data.
Rcpp::StringMatrix myMatrix( j-i , myData.ncol());
Rcpp::colnames(myMatrix) = Rcpp::colnames(myData);
// Populate the return matrix
for(int k = 0; k < myMatrix.nrow(); k++){
Rcpp::checkUserInterrupt();
myMatrix(k, Rcpp::_) = myData(k+i, Rcpp::_);
}
return myMatrix;
} else {
// Fill missing data.
Rcpp::StringMatrix myMatrix( end - start + 1 , myData.ncol());
Rcpp::colnames(myMatrix) = Rcpp::colnames(myData);
// Populate the return matrix
if( i >= myData.nrow()){
// No data
for(int k = 0; k < myMatrix.nrow(); k++){
Rcpp::checkUserInterrupt();
myMatrix(k,0) = myBed(0);
// myMatrix(k,1) = std::to_string(start + k);
std::ostringstream stm;
stm << start + k;
myMatrix(k,1) = stm.str();
myMatrix(k,2) = NA_STRING;
// for(int m=2; m<myMatrix.ncol(); m++){
// myMatrix(k,m) = NA_STRING;
// }
}
} else {
// Data and possibly missing data
int l = 0;
for(int k = 0; k < myMatrix.nrow(); k++){
Rcpp::checkUserInterrupt();
if( i + l < myData.nrow() ){
// We have not overrun the file yet
temp = Rcpp::as< std::string >( myData( i+l , 1 ) );
int myPOS = atoi(temp.c_str());
// int myPOS = stoi(temp);
if( myPOS == start + k ){
// myMatrix(k, Rcpp::_) = myData(k+i, Rcpp::_);
myMatrix(k, Rcpp::_) = myData( i + l, Rcpp::_);
l++;
} else {
myMatrix(k,0) = myBed(0);
// myMatrix(k,1) = std::to_string(myPOS);
std::ostringstream stm;
stm << myPOS;
myMatrix(k,1) = stm.str();
myMatrix(k,2) = NA_STRING;
//myMatrix(k,1) = myBed(1) + k;
//for(int m=2; m<myMatrix.ncol(); m++){
// myMatrix(k,m) = NA_STRING;
//}
}
} else {
// We've overrun the rows in the file.
myMatrix(k,0) = myBed(0);
// myMatrix(k,1) = std::to_string( start + k );
std::ostringstream stm;
stm << start + k;
myMatrix(k,1) = stm.str();
myMatrix(k,2) = NA_STRING;
//myMatrix(k,1) = myBed(1) + k;
//for(int m=2; m<myMatrix.ncol(); m++){
// myMatrix(k,m) = NA_STRING;
//}
}
}
}
return myMatrix;
}
Rcpp::Rcerr << "You should never get here, something bad has happened!\n";
}
//' @rdname is_het
//' @name is_het
//'
//'
//'
//' @export
// [[Rcpp::export]]
Rcpp::LogicalMatrix is_het(Rcpp::StringMatrix x,
Rcpp::LogicalVector na_is_false = true
){
// NA matrix to return in case of unexpected results.
// Rcpp::LogicalMatrix nam( 1, 1 );
// nam(0,0) = NA_LOGICAL;
// Initialize return data matrix.
Rcpp::LogicalMatrix hets( x.nrow(), x.ncol() );
hets.attr("dimnames") = x.attr("dimnames");
int i;
int j;
int k;
for( i=0; i<x.nrow(); i++){
for( j=0; j<x.ncol(); j++){
// Parse genotype string into alleles.
std::string my_string;
if( x(i,j) == NA_STRING ){
my_string = ".";
} else {
my_string = x(i,j);
}
std::vector < std::string > allele_vec;
// vcfRCommon::strsplit(my_string, allele_vec, my_split);
int unphased_as_na = 0; // 0 == FALSE
vcfRCommon::gtsplit( my_string, allele_vec, unphased_as_na );
// Rcpp::Rcout << "gtsplit returned: " << allele_vec[0];
// for( k=1; k<allele_vec.size(); k++){
// Rcpp::Rcout << "," << allele_vec[k];
// }
// Rcpp::Rcout << "\n";
// Initialize new vector of alleles with first element of allele_vec.
std::vector < std::string > allele_vec2;
// Scroll through vector looking for alleles.
for(k=0; k<allele_vec.size(); k++){
if( allele_vec[k] == "." ){
// Found missing value.
// Delete and bail out.
while( allele_vec2.size() > 0 ){
allele_vec2.erase( allele_vec2.begin() );
}
k = allele_vec.size();
} else if( allele_vec2.size() == 0 ){
// Initialize.
allele_vec2.push_back( allele_vec[k] );
} else if( allele_vec2[0] != allele_vec[k] ){
allele_vec2.push_back( allele_vec[k] );
}
}
// Rcpp::Rcout << "allele_vec2.size(): " << allele_vec2.size();
// Rcpp::Rcout << "\n";
// Rcpp::Rcout << "\n";
// Score return value.
if( allele_vec2.size() == 0){
if( na_is_false[0] == true ){
hets(i,j) = false;
} else if( na_is_false[0] == false ){
hets(i,j) = NA_LOGICAL;
}
} else if( allele_vec2.size() == 1){
hets(i,j) = false;
} else if( allele_vec2.size() > 1){
hets(i,j) = true;
}
}
}
return( hets );
}