void Motif::freq_matrix_extended(double *fm) const{
const vector<vector <int> >& seq = seqset.seq();
int i, col, j;
int fm_size = (width + 2 * ncols()) * 4;
for(i = 0; i < fm_size; i++) fm[i] = 0.0;
if(number() == 0) return;
for(i = 0; i < number(); i++) {//i = site number
int c = chrom(i);
int p = posit(i);
bool s = strand(i);
for(j = 0, col = -ncols(); col < width + ncols(); col++, j += 4) {
if(s) {
if((p + col <= seqset.len_seq(c) - 1) && (p + col >= 0)) {
assert(j + seq[c][p + col] < fm_size);
fm[j + seq[c][p + col]] += 1.0;
} else {
for(int k = 0; k < 4; k++) {
fm[j + k] += 0.25;
}
}
} else {
if((p + width - 1 - col <= seqset.len_seq(c) - 1) && (p + width - 1 - col >= 0)) {
assert(j + 3 - seq[c][p + width - 1 - col] < fm_size);
fm[j + 3 - seq[c][p + width - 1 - col]] += 1.0;
} else {
for(int k = 0; k < 4; k++) {
fm[j + k] += 0.25;
}
}
}
}
}
for(i = 0; i < fm_size; i++) fm[i] /= (double) number();
}
int main()
{
const unsigned POP_SIZE = 8, CHROM_SIZE = 16;
unsigned i;
// a chromosome randomizer
eoInitPermutation <Chrom> random(CHROM_SIZE);
// the population:
eoPop<Chrom> pop;
// Evaluation
eoEvalFuncPtr<Chrom> eval( real_value );
for (i = 0; i < POP_SIZE; ++i)
{
Chrom chrom(CHROM_SIZE);
std::cout << " Initial chromosome n°" << i << " : " << chrom << "..." << std::endl;
random(chrom);
eval(chrom);
std::cout << " ... becomes : " << chrom << " after initialization" << std::endl;
check_permutation(chrom);
pop.push_back(chrom);
}
return 0;
}
void
site_crawler::
dump_state(std::ostream& os) const {
const std::string& afile(_si.file);
os << "LOCUS_CRAWLER STATE:\n";
os << "\tchrom: " << chrom();
os << "\tposition: " << pos() << " offset: " << _locus_offset << "\n";
os << "\tis_indel: " << is_indel() << "\n";
os << "\tfile: '" << afile << "'\n";
os << "\tline: '";
dump_line(os);
os << "'\n";
}
main()
{
Pop pop;
EORandom<float> random;
Chrom chrom(NUM_GENES, BITSxGEN);
for (unsigned i = 0; i < POP_SIZE; i++)
{
random(chrom);
chrom.fitness(i);
pop.push_back(chrom);
}
Bin fitness;
EOEvalAll<Chrom> eval(fitness);
EOStat<Chrom> graphEval( eval, true ); // True is for verbose
EORank<Chrom> transform(1);
// EOMutation<Chrom> mutation(0.05);
// transform.addOp((EOOp<Chrom>*)&mutation);
NxOver<Chrom> nxover(2);
transform.addOp(&nxover);
EOGenTerm<Chrom> term(100);
EOAgeGA<Chrom> agega(graphEval, transform, term);
try
{
agega(pop);
}
catch(UException& e)
{
cout << e.what() << endl;
}
cout << "the best: " << pop.front() << endl;
return 0;
}
main()
{
const unsigned POP_SIZE = 8, CHROM_SIZE = 4;
unsigned i;
eoBinRandom<Chrom> random;
eoPop<Chrom> pop;
// Create the population
for (i = 0; i < POP_SIZE; ++i) {
Chrom chrom(CHROM_SIZE);
random(chrom);
BinaryValue()(chrom);
pop.push_back(chrom);
}
// print population
std::cout << "population:" << std::endl;
for (i = 0; i < pop.size(); ++i)
std::cout << pop[i] << " " << pop[i].fitness() << std::endl;
// Declare 1-selectors
eoUniformSelect<Chrom> uSelect;
Chrom aChrom;
aChrom = uSelect( pop );
std::cout << "Uniform Select " << aChrom << " " << aChrom.fitness() << std::endl;
eoStochTournament<Chrom> sSelect(0.7);
aChrom = sSelect( pop );
std::cout << "Stochastic Tournament " << aChrom << " " << aChrom.fitness() << std::endl;
eoDetTournament<Chrom> dSelect(3);
aChrom = dSelect( pop );
std::cout << "Deterministic Tournament " << aChrom << " " << aChrom.fitness() << std::endl;
return 0;
}
void Fitness_test::testFitness()
{
unsigned int aPrecision = 4;
std::vector<double> adomain;
adomain.push_back(1.0);
adomain.push_back(2.0);
adomain.push_back(3.0);
adomain.push_back(6.0);
adomain.push_back(8.0);
adomain.push_back(16.0);
std::vector<unsigned int> length;
unsigned int aNDIM = 3;
Chromosome_Length chrom(aPrecision, aNDIM, adomain);
chrom.buildChromosome();
length = chrom.getLength();
Chromosome achromosome = chrom.getChromosome();
for(unsigned int i=0; i<achromosome.size(); i++)
{
if(i%2==0)
{
achromosome[i]=true;
}
else
achromosome[i]=false;
}
Function afunction;
Fittness_value<Function> fitness(length, aNDIM, adomain, afunction);
double result;
result = fitness.computeValue(achromosome);
std:: cout << "the final value is = " << result << std::endl;
}
int main()
{
const unsigned POP_SIZE = 3, CHROM_SIZE = 8;
unsigned i;
// a chromosome randomizer
eoInitPermutation <Chrom> random(CHROM_SIZE);
// the population:
eoPop<Chrom> pop;
// Evaluation
//eoEvalFuncPtr<Chrom> eval( real_value );
for (i = 0; i < POP_SIZE; ++i)
{
Chrom chrom(CHROM_SIZE);
random(chrom);
//eval(chrom);
pop.push_back(chrom);
}
// a shift mutation
eoOrderXover<Chrom> cross;
for (i = 0; i < POP_SIZE; ++i)
std::cout << " Initial chromosome n�" << i << " : " << pop[i] << "..." << std::endl;
cross(pop[0],pop[1]);
cross(pop[1],pop[2]);
for (i = 0; i < POP_SIZE; ++i) {
std::cout << " Initial chromosome n�" << i << " becomes : " << pop[i] << " after orderXover" << std::endl;
check_permutation(pop[i]);
}
return 0;
}
int main()
{
const unsigned SIZE = 8;
eoBooleanGenerator gen;
eo::rng.reseed( time( 0 ) );
Chrom chrom(SIZE), chrom2(SIZE);
chrom.fitness(binary_value(chrom)); chrom2.fitness(binary_value(chrom2));
std::cout << chrom << std::endl;
std::cout << chrom2 << std::endl;
// Virus Mutation
VirusBitFlip<float> vf;
unsigned i;
for ( i = 0; i < 10; i++ ) {
vf( chrom );
std::cout << chrom << std::endl;
}
// Chrom Mutation
std::cout << "Chrom mutation--------" << std::endl;
VirusMutation<float> vm;
for ( i = 0; i < 10; i++ ) {
vm( chrom );
std::cout << chrom << std::endl;
}
// Chrom Transmision
std::cout << "Chrom transmission--------" << std::endl;
VirusTransmission<float> vt;
vt( chrom2, chrom );
std::cout << chrom2 << std::endl;
return 0;
}
main()
{
const unsigned POP_SIZE = 8, CHROM_SIZE = 16;
unsigned i;
// a chromosome randomizer
eoBinRandom<Chrom> random;
// the populations:
eoPop<Chrom> pop;
// Evaluation
eoEvalFuncPtr<Chrom> eval( binary_value );
for (i = 0; i < POP_SIZE; ++i)
{
Chrom chrom(CHROM_SIZE);
random(chrom);
eval(chrom);
pop.push_back(chrom);
}
std::cout << "population:" << std::endl;
for (i = 0; i < pop.size(); ++i)
std::cout << "\t" << pop[i] << " " << pop[i].fitness() << std::endl;
// selection
eoLottery<Chrom> lottery;
// breeder
eoBinBitFlip<Chrom> bitflip;
eoBinCrossover<Chrom> xover;
eoProportionalOpSel<Chrom> propSel;
eoBreeder<Chrom> breeder( propSel );
propSel.addOp(bitflip, 0.25);
propSel.addOp(xover, 0.75);
// replacement
eoInclusion<Chrom> inclusion;
// Terminators
eoFitTerm<Chrom> term( pow(2.0, CHROM_SIZE), 1 );
// GA generation
eoEasyEA<Chrom> ea(lottery, breeder, inclusion, eval, term);
// evolution
try
{
ea(pop);
}
catch (std::exception& e)
{
std::cout << "exception: " << e.what() << std::endl;;
exit(EXIT_FAILURE);
}
std::cout << "pop" << std::endl;
for (i = 0; i < pop.size(); ++i)
std::cout << "\t" << pop[i] << " " << pop[i].fitness() << std::endl;
return 0;
}
void main_function()
{
const unsigned SIZE = 8;
unsigned i, j;
eoBooleanGenerator gen;
Chrom chrom(SIZE), chrom2;
chrom.fitness(binary_value(chrom)); chrom2.fitness(binary_value(chrom2));
std::cout << "chrom: " << chrom << std::endl;
chrom[0] = chrom[SIZE - 1] = true; chrom.fitness(binary_value(chrom));
std::cout << "chrom: " << chrom << std::endl;
chrom[0] = chrom[SIZE - 1] = false; chrom.fitness(binary_value(chrom));
std::cout << "chrom: " << chrom << std::endl;
chrom[0] = chrom[SIZE - 1] = true; chrom.fitness(binary_value(chrom));
std::cout << "chrom.className() = " << chrom.className() << std::endl;
std::cout << "chrom: " << chrom << std::endl
<< "chrom2: " << chrom2 << std::endl;
std::ostringstream os;
os << chrom;
std::istringstream is(os.str());
is >> chrom2; chrom.fitness(binary_value(chrom2));
std::cout << "\nTesting reading, writing\n";
std::cout << "chrom: " << chrom << "\nchrom2: " << chrom2 << '\n';
std::fill(chrom.begin(), chrom.end(), false);
std::cout << "--------------------------------------------------"
<< std::endl << "eoMonOp's aplied to .......... " << chrom << std::endl;
eoInitFixedLength<Chrom>
random(chrom.size(), gen);
random(chrom); chrom.fitness(binary_value(chrom));
std::cout << "after eoBinRandom ............ " << chrom << std::endl;
eoOneBitFlip<Chrom> bitflip;
bitflip(chrom); chrom.fitness(binary_value(chrom));
std::cout << "after eoBitFlip .............. " << chrom << std::endl;
eoBitMutation<Chrom> mutation(0.5);
mutation(chrom); chrom.fitness(binary_value(chrom));
std::cout << "after eoBinMutation(0.5) ..... " << chrom << std::endl;
eoBitInversion<Chrom> inversion;
inversion(chrom); chrom.fitness(binary_value(chrom));
std::cout << "after eoBinInversion ......... " << chrom << std::endl;
eoBitNext<Chrom> next;
next(chrom); chrom.fitness(binary_value(chrom));
std::cout << "after eoBinNext .............. " << chrom << std::endl;
eoBitPrev<Chrom> prev;
prev(chrom); chrom.fitness(binary_value(chrom));
std::cout << "after eoBinPrev .............. " << chrom << std::endl;
std::fill(chrom.begin(), chrom.end(), false); chrom.fitness(binary_value(chrom));
std::fill(chrom2.begin(), chrom2.end(), true); chrom2.fitness(binary_value(chrom2));
std::cout << "--------------------------------------------------"
<< std::endl << "eoBinOp's aplied to ... "
<< chrom << " " << chrom2 << std::endl;
eo1PtBitXover<Chrom> xover;
std::fill(chrom.begin(), chrom.end(), false);
std::fill(chrom2.begin(), chrom2.end(), true);
xover(chrom, chrom2);
chrom.fitness(binary_value(chrom)); chrom2.fitness(binary_value(chrom2));
std::cout << "eoBinCrossover ........ " << chrom << " " << chrom2 << std::endl;
for (i = 1; i < SIZE; i++)
{
eoNPtsBitXover<Chrom> nxover(i);
std::fill(chrom.begin(), chrom.end(), false);
std::fill(chrom2.begin(), chrom2.end(), true);
nxover(chrom, chrom2);
chrom.fitness(binary_value(chrom)); chrom2.fitness(binary_value(chrom2));
std::cout << "eoBinNxOver(" << i << ") ........ "
<< chrom << " " << chrom2 << std::endl;
}
for (i = 1; i < SIZE / 2; i++)
for (j = 1; j < SIZE / 2; j++)
{
eoBitGxOver<Chrom> gxover(i, j);
std::fill(chrom.begin(), chrom.end(), false);
std::fill(chrom2.begin(), chrom2.end(), true);
gxover(chrom, chrom2);
chrom.fitness(binary_value(chrom)); chrom2.fitness(binary_value(chrom2));
std::cout << "eoBinGxOver(" << i << ", " << j << ") ..... "
<< chrom << " " << chrom2 << std::endl;
}
// test SGA algorithm
eoGenContinue<Chrom> continuator1(50);
eoFitContinue<Chrom> continuator2(65535.f);
eoCombinedContinue<Chrom> continuator(continuator1, continuator2);
eoCheckPoint<Chrom> checkpoint(continuator);
eoStdoutMonitor monitor;
checkpoint.add(monitor);
eoSecondMomentStats<Chrom> stats;
monitor.add(stats);
checkpoint.add(stats);
eoProportionalSelect<Chrom> select;
eoEvalFuncPtr<Chrom> eval(binary_value);
eoSGA<Chrom> sga(select, xover, 0.8f, bitflip, 0.1f, eval, checkpoint);
eoInitFixedLength<Chrom> init(16, gen);
eoPop<Chrom> pop(100, init);
apply<Chrom>(eval, pop);
sga(pop);
pop.sort();
std::cout << "Population " << pop << std::endl;
std::cout << "\nBest: " << pop[0].fitness() << '\n';
/*
Commented this out, waiting for a definite decision what to do with the mOp's
// Check multiOps
eoMultiMonOp<Chrom> mOp( &next );
mOp.adOp( &bitflip );
std::cout << "before multiMonOp............ " << chrom << std::endl;
mOp( chrom );
std::cout << "after multiMonOp .............. " << chrom << std::endl;
eoBinGxOver<Chrom> gxover(2, 4);
eoMultiBinOp<Chrom> mbOp( &gxover );
mOp.adOp( &bitflip );
std::cout << "before multiBinOp............ " << chrom << " " << chrom2 << std::endl;
mbOp( chrom, chrom2 );
std::cout << "after multiBinOp .............. " << chrom << " " << chrom2 <<std::endl;
*/
}
// [[Rcpp::export]]
Rcpp::DataFrame vcf_body(std::string x, Rcpp::NumericVector stats) {
// Read in the fixed and genotype portion of the file.
// Stats contains:
// "meta", "header", "variants", "columns"
Rcpp::CharacterVector chrom(stats[2]);
Rcpp::IntegerVector pos(stats[2]);
Rcpp::StringVector id(stats[2]);
Rcpp::StringVector ref(stats[2]);
Rcpp::StringVector alt(stats[2]);
Rcpp::NumericVector qual(stats[2]);
Rcpp::StringVector filter(stats[2]);
Rcpp::StringVector info(stats[2]);
// if(stats[3])
Rcpp::CharacterMatrix gt(stats[2], stats[3] - 8);
std::string line; // String for reading file into
// Open file.
std::ifstream myfile;
myfile.open (x.c_str(), std::ios::in);
if (!myfile.is_open()){
Rcpp::Rcout << "Unable to open file";
}
// Iterate past meta.
int i = 0;
int j = 0;
while ( i < stats[0] ){
getline (myfile,line);
i++;
}
// Get header.
getline (myfile,line);
std::string header = line;
// Get body.
i = 0;
char buffer [50];
while ( getline (myfile,line) ){
Rcpp::checkUserInterrupt();
std::vector < std::string > temps = tabsplit(line, stats[3]);
if(temps[0] == "."){
chrom[i] = NA_STRING;
} else {
chrom[i] = temps[0];
}
if(temps[1] == "."){
pos[i] = NA_INTEGER;
} else {
pos[i] = atoi(temps[1].c_str());
}
if(temps[2] == "."){
id[i] = NA_STRING;
} else {
id[i] = temps[2];
}
if(temps[3] == "."){
ref[i] = NA_STRING;
} else {
ref[i] = temps[3];
}
if(temps[4] == "."){
alt[i] = NA_STRING;
} else {
alt[i] = temps[4];
}
if(temps[5] == "."){
qual[i] = NA_REAL;
} else {
qual[i] = atof(temps[5].c_str());
}
if(temps[6] == "."){
filter[i] = NA_STRING;
} else {
filter[i] = temps[6];
}
if(temps[7] == "."){
info[i] = NA_STRING;
} else {
info[i] = temps[7];
}
for(j=8; j<stats[3]; j++){
gt(i, j-8) = temps[j];
// body(i, j-8) = temps[j];
}
i++;
if( i % nreport == 0){
Rcpp::Rcout << "\rProcessed variant: " << i;
}
}
myfile.close();
Rcpp::Rcout << "\rProcessed variant: " << i;
Rcpp::Rcout << "\nAll variants processed\n";
Rcpp::DataFrame df1 = Rcpp::DataFrame::create(
Rcpp::_["CHROM"]= chrom,
Rcpp::_["POS"]= pos,
Rcpp::_["ID"] = id,
Rcpp::_["REF"] = ref,
Rcpp::_["ALT"] = alt,
Rcpp::_["QUAL"] = qual,
Rcpp::_["FILTER"] = filter,
Rcpp::_["INFO"] = info,
gt);
std::vector < std::string > temps = tabsplit(header, stats[3]);
temps[0].erase(0,1);
df1.names() = temps;
Rcpp::Rcout << "Rcpp::DataFrame created.\n";
return df1;
}
bool positionMatch(const Variant& rhs) const {
return chrom() == rhs.chrom() &&
start() == rhs.start() &&
stop() == rhs.stop();
}
void BedGenomeCoverage::CoverageBam(string bamFile) {
ResetChromCoverage();
// open the BAM file
BamReader reader;
if (!reader.Open(bamFile)) {
cerr << "Failed to open BAM file " << bamFile << endl;
exit(1);
}
// get header & reference information
string header = reader.GetHeaderText();
RefVector refs = reader.GetReferenceData();
// load the BAM header references into a BEDTools "genome file"
_genome = new GenomeFile(refs);
// convert each aligned BAM entry to BED
// and compute coverage on B
BamAlignment bam;
while (reader.GetNextAlignment(bam)) {
// skip if the read is unaligned
if (bam.IsMapped() == false)
continue;
bool _isReverseStrand = bam.IsReverseStrand();
//changing second mate's strand to opposite
if( _dUTP && bam.IsPaired() && bam.IsMateMapped() && bam.IsSecondMate())
_isReverseStrand = !bam.IsReverseStrand();
// skip if we care about strands and the strand isn't what
// the user wanted
if ( (_filterByStrand == true) &&
((_requestedStrand == "-") != _isReverseStrand) )
continue;
// extract the chrom, start and end from the BAM alignment
string chrom(refs.at(bam.RefID).RefName);
CHRPOS start = bam.Position;
CHRPOS end = bam.GetEndPosition(false, false) - 1;
// are we on a new chromosome?
if ( chrom != _currChromName )
StartNewChrom(chrom);
if(_pair_chip_) {
// Skip if not a proper pair
if (bam.IsPaired() && (!bam.IsProperPair() or !bam.IsMateMapped()) )
continue;
// Skip if wrong coordinates
if( ( (bam.Position<bam.MatePosition) && bam.IsReverseStrand() ) ||
( (bam.MatePosition < bam.Position) && bam.IsMateReverseStrand() ) ) {
//chemically designed: left on positive strand, right on reverse one
continue;
}
/*if(_haveSize) {
if (bam.IsFirstMate() && bam.IsReverseStrand()) { //put fragmentSize in to the middle of pair end_fragment
int mid = bam.MatePosition+abs(bam.InsertSize)/2;
if(mid<_fragmentSize/2)
AddCoverage(0, mid+_fragmentSize/2);
else
AddCoverage(mid-_fragmentSize/2, mid+_fragmentSize/2);
}
else if (bam.IsFirstMate() && bam.IsMateReverseStrand()) { //put fragmentSize in to the middle of pair end_fragment
int mid = start+abs(bam.InsertSize)/2;
if(mid<_fragmentSize/2)
AddCoverage(0, mid+_fragmentSize/2);
else
AddCoverage(mid-_fragmentSize/2, mid+_fragmentSize/2);
}
} else */
if (bam.IsFirstMate() && bam.IsReverseStrand()) { //prolong to the mate to the left
AddCoverage(bam.MatePosition, end);
}
else if (bam.IsFirstMate() && bam.IsMateReverseStrand()) { //prolong to the mate to the right
AddCoverage(start, start + abs(bam.InsertSize) - 1);
}
} else if (_haveSize) {
if(bam.IsReverseStrand()) {
if(end<_fragmentSize) { //sometimes fragmentSize is bigger :(
AddCoverage(0, end);
} else {
AddCoverage(end + 1 - _fragmentSize, end );
}
} else {
AddCoverage(start,start+_fragmentSize - 1);
}
} else
// add coverage accordingly.
if (!_only_5p_end && !_only_3p_end) {
bedVector bedBlocks;
// we always want to split blocks when a D CIGAR op is found.
// if the user invokes -split, we want to also split on N ops.
if (_obeySplits) { // "D" true, "N" true
GetBamBlocks(bam, refs.at(bam.RefID).RefName, bedBlocks, true, true);
}
else { // "D" true, "N" false
GetBamBlocks(bam, refs.at(bam.RefID).RefName, bedBlocks, true, false);
}
AddBlockedCoverage(bedBlocks);
}
else if (_only_5p_end) {
CHRPOS pos = ( !bam.IsReverseStrand() ) ? start : end;
AddCoverage(pos,pos);
}
else if (_only_3p_end) {
CHRPOS pos = ( bam.IsReverseStrand() ) ? start : end;
AddCoverage(pos,pos);
}
}
// close the BAM
reader.Close();
// process the results of the last chromosome.
ReportChromCoverage(_currChromCoverage, _currChromSize,
_currChromName, _currChromDepthHist);
// report all empty chromsomes
PrintEmptyChromosomes();
// report the overall coverage if asked.
PrintFinalCoverage();
}
