GenotyperFH Class Reference

FileHandler to write joint genotype files for multiple mappable traits. More...

#include <servicenotifiers.h>

Inheritance diagram for GenotyperFH:

Collaboration diagram for GenotyperFH:

Public Member Functions
	GenotyperFH (LCE_FileServicesNotifier *event)
virtual	~GenotyperFH ()
virtual void	FHwrite ()
virtual void	FHread (string &filename)
void	setTraits (vector< trait_t > &traits)
void	setTraitIndex (vector< unsigned int > &indices)
void	setFormat (string format)
void	setLogtime (string logtime)
void	set_isDiallelic (bool test)
void	set_trimFixedLoci (bool test, double maf)
void	prepare_data_table (map< unsigned int, vector< unsigned int > > &trait_locus, vector< string > &col_names)
void	collate_trait_map_positions (vector< unsigned int > &positions, map< unsigned int, vector< unsigned int > > &table)
void	print_genotypes (ofstream &FH, map< unsigned int, vector< unsigned int > > &table)
void	print_snp_genotypes (ofstream &FH, map< unsigned int, vector< unsigned int > > &table)
void	print_snp_id (ofstream &FH, map< unsigned int, vector< unsigned int > > &table)
void	print_map_positions (ofstream &FH, vector< unsigned int > &positions, vector< string > &col_names)
Public Member Functions inherited from EventFileHandler< LCE_FileServicesNotifier >
	EventFileHandler (LCE_FileServicesNotifier *event, const char *ext)
virtual	~EventFileHandler ()
virtual void	FHread (string &filename)=0
virtual void	set (bool rpl_per, bool gen_per, int rpl_occ, int gen_occ, int rank, string path, LCE_FileServicesNotifier *event)
Public Member Functions inherited from FileHandler
	FileHandler (const char *ext)
virtual	~FileHandler ()
virtual void	init ()
	Called by notifier during simulation setup, performs file checking. More...
virtual vector< string >	ifExist ()
	Checks if any file associated with the current file name already exists on disk. More...
virtual void	set (bool rpl_per, bool gen_per, int rpl_occ, int gen_occ, int rank, string path)
	Sets the hanlder parameters. More...
virtual void	set_multi (bool rpl_per, bool gen_per, int rpl_occ, TMatrix *Occ, string path)
virtual void	update ()
	Updates the inner replicate and generation counters and calls FHwrite if needed by the the periodicity of the file. More...
Metapop *	get_pop_ptr ()
	Returns the pointer to the current metapop through the FileServices interface. More...
void	set_pop_ptr (Metapop *pop_ptr)
FileServices *	get_service ()
	Returns pointer to the FileServices. More...
void	set_service (FileServices *srv)
std::string &	get_path ()
void	set_path ()
std::string &	get_extension ()
void	set_extension (const char *ext)
std::string &	get_filename ()
	Builds and returns the current file name depending on the periodicity of the file. More...
bool	get_isInputHandler ()
void	set_isInputHandler (bool val)
bool	get_isReplicatePeriodic ()
void	set_isReplicatePeriodic (bool val)
unsigned int	get_ReplicateOccurrence ()
void	set_ReplicateOccurrence (unsigned int val)
bool	get_isGenerationPeriodic ()
void	set_isGenerationPeriodic (bool val)
unsigned int	get_GenerationOccurrence ()
void	set_GenerationOccurrence (unsigned int val)
unsigned int	get_ExecRank ()
	unused yet... More...
void	set_ExecRank (int val)
TMatrix *	get_OccMatrix ()
void	set_OccMatrix (TMatrix *occ)
bool	get_isMasterExec ()
void	set_isMasterExec (bool is)
Public Member Functions inherited from Handler
virtual	~Handler ()

Private Attributes
vector< trait_t >	_traits
vector< unsigned int >	_trait_index
string	_format
string	_logtime
bool	_is_diallelic
bool	_trim_loci
double	_trim_maf

Additional Inherited Members
Protected Attributes inherited from EventFileHandler< LCE_FileServicesNotifier >
LCE_FileServicesNotifier *	_FHLinkedEvent
Protected Attributes inherited from FileHandler
Metapop *	_pop
	Pointer to the current metapop, set during initialization within the init function. More...

Detailed Description

FileHandler to write joint genotype files for multiple mappable traits.

Constructor & Destructor Documentation

GenotyperFH()

GenotyperFH::GenotyperFH ( LCE_FileServicesNotifier *  event )

inline

: EventFileHandler(event, "") , _is_diallelic(0), _trim_loci(0), _trim_maf(0) {}

~GenotyperFH()

virtual GenotyperFH::~GenotyperFH ( )

inlinevirtual

{}

Member Function Documentation

collate_trait_map_positions()

void GenotyperFH::collate_trait_map_positions	(	vector< unsigned int > &	positions,
		map< unsigned int, vector< unsigned int > > &	table
	)

{
  TraitPrototype* tproto;
  unsigned int     index;
 
  for(auto trait : _traits) {
 
    tproto = SIMenv::MainSim->get_pop()->getTraitPrototype(trait);
 
    if(!tproto) fatal("Genotyper::could not access trait prototype in (sampled) pop for trait \"%s\"\n", trait.c_str() );
 
    index = tproto->get_index();
    vector<unsigned int> & locus = table[index];
 
    vector< unsigned int > trait_positions = tproto->get_locus_map_positions();
 
    for(unsigned int i = 0; i < locus.size(); ++i)
      positions.push_back(trait_positions[ locus[i] ]);
  }
}

References _traits, fatal(), TraitPrototype::get_index(), TraitPrototype::get_locus_map_positions(), SimRunner::get_pop(), IndFactory::getTraitPrototype(), and SIMenv::MainSim.

Referenced by FHwrite().

Here is the caller graph for this function:

FHread()

virtual void GenotyperFH::FHread ( string &  filename )

inlinevirtual

Implements FileHandler.

{}

FHwrite()

void GenotyperFH::FHwrite ( )

virtual

Implements EventFileHandler< LCE_FileServicesNotifier >.

{
  if(!_pop->isAlive()) return;
 
  // reset the pop ptr from the file services, will be the main metapop without sub sampling
  // or a sub sampled pop otherwise:
 
//  vector< vector<double> > data_table;
  vector< string >        file_header;
  map< unsigned int, vector<unsigned int> > table_trait_locus;
 
  _pop = get_service()->getSampledPop();
 
  // prepare the table receiving the allelic info. Sets the right number of columns and individual info
  prepare_data_table(table_trait_locus, file_header);
 
  // WRITE data to file
  string filename = this->get_filename();
 
  ofstream FILE(filename, ios::out | ios::trunc);
  std::ios_base::sync_with_stdio(false); // better for writing performances
 
  if(!FILE) fatal("Genotyper::could not open genotype output file \"%s\"\n",filename.c_str());
 
  // print the header
  auto concat = [] (string a, string b){return a + " " + b;};
 
  FILE << std::accumulate(file_header.begin()+1, file_header.end(), *file_header.begin(), concat) << endl;
 
  if(_format == "genotype") {
 
    print_genotypes(FILE, table_trait_locus);
 
  } else if (_format == "snp_genotype") {
 
    print_snp_genotypes(FILE, table_trait_locus);
 
  } else {
 
    print_snp_id(FILE, table_trait_locus);
 
  }
 
  FILE.close();
  std::ios_base::sync_with_stdio(true); // reset
 
  // --------------------------------------------------------------------------------------
  // write the map position of the loci to a separate file:
 
  // collect all loci positions in a vector, ordered as the columns of the data table (minus first 5 columns)
  vector< unsigned int > positions;
 
  collate_trait_map_positions(positions, table_trait_locus);
 
  filename = get_path() + this->get_service()->getGenerationReplicateFileName() + ".map";
 
  FILE.open(filename, ios::out | ios::trunc);
  std::ios_base::sync_with_stdio(false); // better for writing performances
 
  if(!FILE) fatal("Genotyper::could not open map positions output file \"%s\"\n",filename.c_str());
 
  assert(positions.size() == file_header.size()-5 || positions.size() == (file_header.size()-5)/2 );
 
  bool increment = 0;
  if(positions.size() == (file_header.size()-5)/2) increment = 1;
  // record map positions in two column: locus_name, position
  // because diploid, two locus names per locus are present for genotypes and sno_ids...
  for(unsigned int i = 0, c = 5; i < positions.size(); ++i, ++c) {
    FILE << file_header[c] << " " << positions[i] << "\n";
    c += increment;
  }
 
  FILE.close();
  std::ios_base::sync_with_stdio(true); // reset
 
  // reset the pop ptr to the main pop
  _pop = get_service()->get_pop_ptr();
 
}

References _format, FileHandler::_pop, collate_trait_map_positions(), fatal(), FileHandler::get_filename(), FileHandler::get_path(), FileServices::get_pop_ptr(), FileHandler::get_service(), FileServices::getGenerationReplicateFileName(), FileServices::getSampledPop(), Metapop::isAlive(), prepare_data_table(), print_genotypes(), print_snp_genotypes(), and print_snp_id().

prepare_data_table()

void GenotyperFH::prepare_data_table	(	map< unsigned int, vector< unsigned int > > &	trait_locus,
		vector< string > &	col_names
	)

{
  // write each trait separately, first fill-in the big table
  // data table has as many rows as number of individuals in the pop
  // number of columns is equal to tot number of loci + pathc-ID + sex + age + stage + ID
  unsigned int num_locus = 0;
  unsigned int num_ind   = _pop->size();
  unsigned int num_patch = _pop->getPatchNbr();
  TraitPrototype* Tproto = 0;
  TTrait*          trait = 0;
  Patch*           patch = 0;
  age_idx age_classes[2] = {OFFSx, ADLTx};
  unsigned int   trait_index;
  double         divider = 1.0/(num_ind * 2.0);
  double               p = 0;
 
 
//  // add first column for the patch ID
//  ind_data.push_back(vector<double>(num_ind, 0.0));
 
  col_names.clear();
 
  // add header for first column = patch-id
  col_names.push_back("pop");
 
//  // add 4 columns with individual info:
//  for(unsigned int i = 0; i < 4; ++i)
//    ind_data.push_back(vector<double>(num_ind, 0.0));
 
  col_names.push_back("sex");
  col_names.push_back("age");
  col_names.push_back("stage");
  col_names.push_back("ID");
 
  // create columns for number of loci per trait, prune loci with p < maf
  for(auto trait_type : _traits) {
 
    Tproto = SIMenv::MainSim->get_pop()->getTraitPrototype(trait_type);
 
    if(!Tproto) fatal("Genotyper::could not access trait prototype in (sampled) pop for trait \"%s\"\n", trait_type.c_str() );
 
    num_locus = Tproto->get_locus_number();
    trait_index = Tproto->get_index();
 
    // check allele frequencies in the population
    for(unsigned int l = 0; l < num_locus; ++l) {
 
      // compute allele frequency in the whole population
      p = 0;
      for(unsigned int i = 0; i < num_patch; ++ i) {
 
        patch = _pop->getPatch(i);
 
        for(unsigned int a = 0; a < 2; ++a) {
          for(unsigned int s = 0; s < 2; ++s) {
            for(unsigned int j = 0; j < patch->size(sex_t(s),age_idx(age_classes[a])); ++j){
 
              trait = patch->get(sex_t(s), age_idx(age_classes[a]), j)->getTrait(trait_index);
              // count the 1's
              p += trait->get_allele(l, 1);
              p += trait->get_allele(l, 0);
            }
          }
        }
      }
 
      p *= divider;
 
      if(p > _trim_maf && p < 1.0 - _trim_maf) {
 
        // add this locus to the list of polymorphic loci
        trait_locus[trait_index].push_back(l);
 
        // add that locus to the table
        if(_format == "snp_genotype") {
          // write locus data in one column 0/1/2
          col_names.push_back(trait_type + "." + to_string(l+1));
        } else {
          col_names.push_back(trait_type + "." + to_string(l+1) + "x");
          col_names.push_back(trait_type + "." + to_string(l+1) + "y");
        }
      }
    }
  }
 
 
}

References _format, FileHandler::_pop, _traits, _trim_maf, ADLTx, fatal(), Patch::get(), TTrait::get_allele(), TraitPrototype::get_index(), TraitPrototype::get_locus_number(), SimRunner::get_pop(), Metapop::getPatch(), Metapop::getPatchNbr(), Individual::getTrait(), IndFactory::getTraitPrototype(), SIMenv::MainSim, OFFSx, Metapop::size(), and Patch::size().

Referenced by FHwrite().

Here is the caller graph for this function:

print_genotypes()

void GenotyperFH::print_genotypes	(	ofstream &	FH,
		map< unsigned int, vector< unsigned int > > &	table
	)

{
//cout<< ">>>>>> GenotyperFH::fill_genotypes\n";
 
  unsigned int num_patch = _pop->getPatchNbr();
  TTrait*          trait = 0;
  Patch*           patch;
  Individual*        ind;
  age_idx age_classes[2] = {OFFSx, ADLTx};
 
  for(unsigned int i = 0; i < num_patch; ++ i) {
 
    patch = _pop->getPatch(i);
 
    for(unsigned int a = 0; a < 2; ++a) {
 
      for(unsigned int s = 0; s < 2; ++s) {
 
        for(unsigned int j = 0; j < patch->size(sex_t(s),age_idx(age_classes[a])); ++j){
 
          ind = patch->get(sex_t(s), age_idx(age_classes[a]), j);
 
          FH << patch->getID()+1 << " " << ind->getSex() << " " << ind->getAge() << " " << age_classes[a] << " " <<ind->getID();
 
          for(auto trait_index : _trait_index) {
 
            trait = ind->getTrait(trait_index);
 
            vector<unsigned int> & loc_list = table[trait_index];
 
            // write locus genotypes
            for(unsigned int l = 0; l < loc_list.size(); ++l) {
              FH << " " << trait->get_allele_value(loc_list[l], FEM);
              FH << " " << trait->get_allele_value(loc_list[l], MAL);
            } // end of loc loop
          } // end for traits
          FH << "\n";
        } // end ind loop
      } // end s loop
    } // end a loop
  } // end patch loop
 
}

References FileHandler::_pop, _trait_index, ADLTx, FEM, Patch::get(), TTrait::get_allele_value(), Individual::getAge(), Individual::getID(), Patch::getID(), Metapop::getPatch(), Metapop::getPatchNbr(), Individual::getSex(), Individual::getTrait(), MAL, OFFSx, and Patch::size().

Referenced by FHwrite().

Here is the caller graph for this function:

print_map_positions()

void GenotyperFH::print_map_positions	(	ofstream &	FH,
		vector< unsigned int > &	positions,
		vector< string > &	col_names
	)

print_snp_genotypes()

void GenotyperFH::print_snp_genotypes	(	ofstream &	FH,
		map< unsigned int, vector< unsigned int > > &	table
	)

{
//  cout<< ">>>>>> GenotyperFH::fill_snp_genotypes\n";
  unsigned int num_patch = _pop->getPatchNbr();
  TTrait*          trait = 0;
  Patch*           patch;
  Individual*        ind;
  age_idx age_classes[2] = {OFFSx, ADLTx};
 
  for(unsigned int i = 0; i < num_patch; ++ i) {
 
    patch = _pop->getPatch(i);
 
    for(unsigned int a = 0; a < 2; ++a) {
 
      for(unsigned int s = 0; s < 2; ++s) {
 
        for(unsigned int j = 0; j < patch->size(sex_t(s),age_idx(age_classes[a])); ++j){
 
          ind = patch->get(sex_t(s), age_idx(age_classes[a]), j);
 
          FH << patch->getID()+1 << " " << ind->getSex() << " " << ind->getAge() << " " << age_classes[a] << " " <<ind->getID();
 
          for(auto trait_index : _trait_index) {
 
            trait = ind->getTrait(trait_index);
 
            vector<unsigned int> & loc_list = table[trait_index];
 
            // write locus snp genotype 0/1/2
            for(unsigned int l = 0, a, b; l < loc_list.size(); ++l) {
              a = trait->get_allele(loc_list[l], FEM);
              b = trait->get_allele(loc_list[l], MAL);
              FH << " " << a + b;
            } // end of loc loop
          } // end for traits
          FH << "\n";
        } // end ind loop
      } // end s loop
    } // end a loop
  } // end patch loop
 
}

References FileHandler::_pop, _trait_index, ADLTx, FEM, Patch::get(), TTrait::get_allele(), Individual::getAge(), Individual::getID(), Patch::getID(), Metapop::getPatch(), Metapop::getPatchNbr(), Individual::getSex(), Individual::getTrait(), MAL, OFFSx, and Patch::size().

Referenced by FHwrite().

Here is the caller graph for this function:

print_snp_id()

void GenotyperFH::print_snp_id	(	ofstream &	FH,
		map< unsigned int, vector< unsigned int > > &	table
	)

{
//  cout<< ">>>>>> GenotyperFH::fill_snp_id\n";
  unsigned int num_patch = _pop->getPatchNbr();
  TTrait*          trait = 0;
  Patch*     patch;
  Individual*  ind;
  age_idx age_classes[2] = {OFFSx, ADLTx};
 
  for(unsigned int i = 0; i < num_patch; ++ i) {
 
    patch = _pop->getPatch(i);
 
    for(unsigned int a = 0; a < 2; ++a) {
 
      for(unsigned int s = 0; s < 2; ++s) {
 
        for(unsigned int j = 0; j < patch->size(sex_t(s),age_idx(age_classes[a])); ++j){
 
          ind = patch->get(sex_t(s), age_idx(age_classes[a]), j);
 
          FH << patch->getID()+1 << " " << ind->getSex() << " " << ind->getAge() << " " << age_classes[a] << " " <<ind->getID();
 
          for(auto trait_index : _trait_index) {
 
            trait = ind->getTrait(trait_index);
 
            vector<unsigned int> & loc_list = table[trait_index];
 
            // write locus genotypes
            for(unsigned int l = 0; l < loc_list.size(); ++l) {
              FH << " " << trait->get_allele(loc_list[l], FEM);
              FH << " " << trait->get_allele(loc_list[l], MAL);
            } // end of loc loop
          } // end for traits
          FH << "\n";
        } // end ind loop
      } // end s loop
    } // end a loop
  } // end patch loop
}

References FileHandler::_pop, _trait_index, ADLTx, FEM, Patch::get(), TTrait::get_allele(), Individual::getAge(), Individual::getID(), Patch::getID(), Metapop::getPatch(), Metapop::getPatchNbr(), Individual::getSex(), Individual::getTrait(), MAL, OFFSx, and Patch::size().

Referenced by FHwrite().

Here is the caller graph for this function:

set_isDiallelic()

void GenotyperFH::set_isDiallelic ( bool  test )

inline

{_is_diallelic = test;}

References _is_diallelic.

Referenced by LCE_FileServicesNotifier::setGenotyperParameters().

Here is the caller graph for this function:

set_trimFixedLoci()

void GenotyperFH::set_trimFixedLoci ( bool  test, double  maf  )

inline

{_trim_loci = test; _trim_maf = maf;}

References _trim_loci, and _trim_maf.

Referenced by LCE_FileServicesNotifier::setGenotyperParameters().

Here is the caller graph for this function:

setFormat()

void GenotyperFH::setFormat ( string  format )

inline

{_format = format;}

References _format.

Referenced by LCE_FileServicesNotifier::setGenotyperParameters().

Here is the caller graph for this function:

setLogtime()

void GenotyperFH::setLogtime ( string  logtime )

inline

{_logtime = logtime;}

References _logtime.

setTraitIndex()

void GenotyperFH::setTraitIndex

(

vector< unsigned int > &

indices

)

{
  _trait_index.assign(traits.begin(), traits.end());
}

References _trait_index.

Referenced by LCE_FileServicesNotifier::setGenotyperParameters().

Here is the caller graph for this function:

setTraits()

void GenotyperFH::setTraits

(

vector< trait_t > &

traits

)

{
  _traits.assign(traits.begin(), traits.end());
}

References _traits.

Referenced by LCE_FileServicesNotifier::setGenotyperParameters().

Here is the caller graph for this function:

Member Data Documentation

_format

string GenotyperFH::_format

private

Referenced by FHwrite(), prepare_data_table(), and setFormat().

_is_diallelic

bool GenotyperFH::_is_diallelic

private

Referenced by set_isDiallelic().

_logtime

string GenotyperFH::_logtime

private

Referenced by setLogtime().

_trait_index

vector< unsigned int > GenotyperFH::_trait_index

private

Referenced by print_genotypes(), print_snp_genotypes(), print_snp_id(), and setTraitIndex().

_traits

vector< trait_t > GenotyperFH::_traits

private

Referenced by collate_trait_map_positions(), prepare_data_table(), and setTraits().

_trim_loci

bool GenotyperFH::_trim_loci

private

Referenced by set_trimFixedLoci().

_trim_maf

double GenotyperFH::_trim_maf

private

Referenced by prepare_data_table(), and set_trimFixedLoci().

---

The documentation for this class was generated from the following files:

• servicenotifiers.h
• servicenotifiers.cc