// [[Rcpp::export]]
Rcpp::List subsetCounts(Rcpp::IntegerVector counts, Rcpp::IntegerVector start, Rcpp::IntegerVector width, Rcpp::LogicalVector strand){
if (start.length() != width.length() || start.length() != strand.length()) Rcpp::stop("provided vectors have different lengths...");
int nr = start.length();
int len = counts.length();
int tot = 0;
int* S = start.begin(); int* W = width.begin();
for (int i = 0; i < nr; ++i){
int s = S[i] - 1;
int w = W[i];
if (s < 0) Rcpp::stop("negative start positions are invalid");
if (s + w > len) Rcpp::stop("range exceeds the lengths of the counts vector");
tot += w;
}
Rcpp::IntegerVector res(tot);
Rcpp::IntegerVector nstart(nr);
Rcpp::IntegerVector nend(nr);
int* R = res.begin(); int* C = counts.begin(); int* ST = strand.begin();
int* NS = nstart.begin(); int* NE = nend.begin();
int currpos = 0;
for (int i = 0; i < nr; ++i){
NS[i] = currpos + 1;
int w = W[i];
if (ST[i]) std::copy(C + S[i]-1, C + S[i]-1 + w, R + currpos);
else std::reverse_copy(C + S[i]-1, C + S[i]-1 + w, R + currpos);
currpos += w;
NE[i] = currpos;
}
return List::create(_("counts")=res, _("starts")=nstart, _("ends")=nend);
}
Rcpp::IntegerVector get_which(Rcpp::LogicalVector row) {
int j = 0;
for (int i = 0; i < row.length(); i++)
if (row(i))
j++;
Rcpp::IntegerVector ret(j);
for (int i = 0, j = 0; i < row.length(); i++)
if (row(i))
ret(j++) = i + 1; // R is 1-based
return ret;
}
// [[Rcpp::export]]
Rcpp::LogicalVector doc_validate(XPtrDoc doc, XPtrDoc schema) {
xmlLineNumbersDefault(1);
Rcpp::CharacterVector vec;
xmlSchemaParserCtxtPtr cptr = xmlSchemaNewDocParserCtxt(schema.checked_get());
xmlSchemaSetParserStructuredErrors(cptr, handleSchemaError, &vec);
xmlSchemaPtr sptr = xmlSchemaParse(cptr);
xmlSchemaValidCtxtPtr vptr = xmlSchemaNewValidCtxt(sptr);
xmlSchemaSetValidStructuredErrors(vptr, handleSchemaError, &vec);
Rcpp::LogicalVector out;
out.push_back(0 == xmlSchemaValidateDoc(vptr, doc.checked_get()));
xmlSchemaFreeParserCtxt(cptr);
xmlSchemaFreeValidCtxt(vptr);
xmlSchemaFree(sptr);
out.attr("errors") = vec;
return out;
}
//[[Rcpp::export]]
Rcpp::RawVector toBitVec(Rcpp::LogicalVector indx) {
unsigned nBit = indx.size();
unsigned nByte = ceil(float(nBit)/8);
Rcpp::RawVector bytes(nByte);//default are all 0s
bytes.attr("bitlen") = nBit;
unsigned byteIndex, bitIndex ;
for(unsigned i = 0 ; i < nBit; i++) {
byteIndex = i / 8;
bitIndex = i % 8;
if(indx(i) == 1)
bytes[byteIndex] = bytes[byteIndex] | 1 << bitIndex;
}
return bytes;
}
// [[Rcpp::export]]
Rcpp::DataFrame gt_to_popsum(Rcpp::DataFrame var_info, Rcpp::CharacterMatrix gt) {
// Calculate popgen summaries for the sample.
// var_info should contain columns named 'CHROM', 'POS', 'mask' and possibly others.
Rcpp::LogicalVector mask = var_info["mask"];
Rcpp::IntegerVector nsample(mask.size());
Rcpp::StringVector allele_counts(mask.size());
Rcpp::NumericVector Hes(mask.size());
Rcpp::NumericVector Nes(mask.size());
int i = 0;
int j = 0;
int k = 0;
for(i=0; i < gt.nrow(); i++){ // Iterate over variants (rows)
if(mask[i] == TRUE){
std::vector<int> myalleles (1,0);
for(j=0; j < gt.ncol(); j++){ // Iterate over samples (columns)
if(gt(i, j) != NA_STRING){
nsample[i]++; // Increment sample count.
// Count alleles per sample.
std::vector < int > intv = gtsplit(as<std::string>(gt(i, j)));
for(k=0; k<intv.size(); k++){
while(myalleles.size() - 1 < intv[k]){
// We have more alleles than exist in the vector myalleles.
myalleles.push_back(0);
}
myalleles[intv[k]]++;
}
}
}
// Concatenate allele counts into a comma delimited string.
int n;
char buffer [50];
n = sprintf (buffer, "%d", myalleles[0]);
for(j=1; j < myalleles.size(); j++){
n=sprintf (buffer, "%s,%d", buffer, myalleles[j]);
}
allele_counts[i] = buffer;
// Sum all alleles.
int nalleles = myalleles[0];
for(j=1; j < myalleles.size(); j++){
nalleles = nalleles + myalleles[j];
}
// Stats.
double He = 1;
He = He - pow(double(myalleles[0])/double(nalleles), myalleles.size());
for(j=1; j < myalleles.size(); j++){
He = He - pow(double(myalleles[j])/double(nalleles), myalleles.size());
}
Hes[i] = He;
Nes[i] = 1/(1-He);
} else { // Missing variant (row=NA)
nsample[i] = NA_INTEGER;
}
}
return Rcpp::DataFrame::create(var_info,
_["n"]=nsample,
_["Allele_counts"]=allele_counts,
_["He"]=Hes,
_["Ne"]=Nes
);
}
//' @export
// [[Rcpp::export(name=".gt_to_popsum")]]
Rcpp::DataFrame gt_to_popsum(Rcpp::DataFrame var_info, Rcpp::CharacterMatrix gt) {
// Calculate popgen summaries for the sample.
// var_info should contain columns named 'CHROM', 'POS', 'mask' and possibly others.
Rcpp::LogicalVector mask = var_info["mask"];
Rcpp::IntegerVector nsample(mask.size());
Rcpp::StringVector allele_counts(mask.size());
Rcpp::NumericVector Hes(mask.size());
Rcpp::NumericVector Nes(mask.size());
int i = 0;
int j = 0;
// unsigned int j = 0;
unsigned int k = 0;
for(i=0; i < gt.nrow(); i++){ // Iterate over variants (rows)
if(mask[i] == TRUE){
std::vector<int> myalleles (1,0);
for(j=0; j < gt.ncol(); j++){ // Iterate over samples (columns)
if(gt(i, j) != NA_STRING){
nsample[i]++; // Increment sample count.
// Rcout << "gt: " << gt(i, j) << "\n";
// Count alleles per sample.
int unphased_as_na = 0; // 0 == FALSE
std::vector < std::string > gt_vector;
std::string gt2 = as<std::string>(gt(i,j));
vcfRCommon::gtsplit( gt2, gt_vector, unphased_as_na );
// Rcout << "gt_vector.size: " << gt_vector.size() << "\n";
for(k=0; k<gt_vector.size(); k++){
int myAllele = std::stoi(gt_vector[k]);
// Rcout << " " << myAllele;
// // If this genotype had an allele we did not previously observe
// we'll have to grow the vector.
while(myalleles.size() - 1 < myAllele){
myalleles.push_back(0);
}
myalleles[myAllele]++;
}
// Rcout << "\n\n";
}
}
// Concatenate allele counts into a comma delimited string.
char buffer [50];
// int n;
// n=sprintf(buffer, "%d", myalleles[0]);
sprintf(buffer, "%d", myalleles[0]);
for(j=1; (unsigned)j < myalleles.size(); j++){
// n=sprintf (buffer, "%s,%d", buffer, myalleles[j]);
sprintf (buffer, "%s,%d", buffer, myalleles[j]);
}
allele_counts[i] = buffer;
// Sum all alleles.
int nalleles = myalleles[0];
for(j=1; (unsigned)j < myalleles.size(); j++){
nalleles = nalleles + myalleles[j];
}
// Stats.
double He = 1;
He = He - pow(double(myalleles[0])/double(nalleles), myalleles.size());
for(j=1; (unsigned)j < myalleles.size(); j++){
He = He - pow(double(myalleles[j])/double(nalleles), myalleles.size());
}
Hes[i] = He;
Nes[i] = 1/(1-He);
} else { // Missing variant (row=NA)
nsample[i] = NA_INTEGER;
}
}
return Rcpp::DataFrame::create(var_info,
_["n"]=nsample,
_["Allele_counts"]=allele_counts,
_["He"]=Hes,
_["Ne"]=Nes
);
}
