bool InputReader::OpenInputFile(std::string p) {
std::vector<std::vector<int>*> left_matrix;
std::vector<std::vector<int>*> right_matrix;
unsigned int rows = 0;
unsigned int cols = 0;
/* MATRICE SINISTRA */
std::ifstream left_file((p+"/left_side").c_str());
/* is there the left matrix? */
if (!left_file.is_open()) {
perror("ERROR: cannot open input file");
printf("%s\n", (p+"/left_side").c_str());
return false;
};
/* read matrix */
std::vector<int> totals_left;
while (left_file.good()) {
std::string riga ;
std::getline(left_file, riga);
unsigned int total_left = 0;
std::vector<int>* ret = split_string_integers(riga, '\t', &total_left);
totals_left.push_back(total_left);
left_matrix.push_back(ret);
};
left_file.close();
rows = left_matrix.size();
cols = left_matrix.back()->size();
/* Initialize compressed ODE matrix */
for ( unsigned int i=0; i<cols; i++ ) {
this->comp_ODE.push_back( new std::vector<int> );
}
/*
RE-parsing matrice sinistra
*/
for (unsigned int r=0; r<rows; r++) {
for (unsigned int c=0; c<cols; c++) {
switch ( left_matrix[r]->at(c) ) {
case 0:
break;
case 1:
this->comp_ODE[ c ]->push_back( totals_left[ r ] +1 );
this->comp_ODE[ c ]->push_back( -(r+1) );
for ( unsigned int cl=0; cl<cols; cl++ ) {
switch ( left_matrix[r]->at(cl) ) {
case 0:
break;
case 1:
this->comp_ODE[ c ]->push_back( cl );
break;
case 2:
this->comp_ODE[ c ]->push_back( cl );
this->comp_ODE[ c ]->push_back( cl );
break;
default:
perror("ERROR: unsupported case");
break;
}
}
break;
default:
perror("ERROR: unsupported case");
break;
}
}
}
/*
MATRICE DESTRA
La matrice destra genera direttamente i prodotti delle reazioni,
compaiono tutti con una costante positiva e possono essere aggiunti on-fly.
*/
std::ifstream right_file((p+"/right_side").c_str());
/* is there the right matrix? */
if (!right_file.is_open()) {
perror("ERROR: cannot open input file");
printf("%s\n", (p+"/right_side").c_str());
return false;
};
/* read matrix */
unsigned int total_right = 0;
while (right_file.good()) {
std::string riga ;
std::getline(right_file, riga);
std::vector<int>* ret = split_string_integers(riga, '\t', &total_right);
right_matrix.push_back(ret);
/*
ret is a single row of the stoichometric matrix.
we parse it and put the corresponding value into the compressed matrix.
*/
for ( unsigned c=0; c<cols; c++ ) {
switch ( ret->at(c) ) {
case 0:
break;
case 1:
/*
We have counted how many stoichiometric values are different from 0
in the left matrix. We use this information to fill the first
value of the compressed matrix. We make +1 because we also consider
the (unknown) kinetic constant.
*/
this->comp_ODE[ c ]->push_back( totals_left[ right_matrix.size()-1 ] +1 );
this->comp_ODE[ c ]->push_back( right_matrix.size() );
for (unsigned int cl=0; cl<cols; cl++) {
switch ( left_matrix[ right_matrix.size()-1 ]->at(cl) ) {
case 0:
break;
case 1:
this->comp_ODE[ c ]->push_back( cl );
break;
case 2:
this->comp_ODE[ c ]->push_back( cl );
this->comp_ODE[ c ]->push_back( cl );
break;
default:
perror("ERROR: unsupported case");
break;
} // end switch
} // end for
break;
case 2:
perror("ERROR: unsupported case");
break;
default:
perror("ERROR: unsupported case");
break;
} // end switch
} // end for
}; // end while
/* Terminators */
for ( unsigned int i=0; i<cols; i++)
this->comp_ODE[ i ]->push_back( 0 );
right_file.close();
return true;
}
void SRAssembler::do_preprocessing(int lib_idx, int file_part){
Library lib = this->libraries[lib_idx];
logger->info("preprocessing lib " + int2str(lib_idx + 1) + ", reads file (" + int2str(file_part) + "/" + int2str(lib.get_num_parts()) + ")");
char suffixc[3];
suffixc[0] = (char)(((file_part-1) / 26) + 97);
suffixc[1] = (char)(((file_part-1) % 26) + 97);
suffixc[2] = '\0';
string suffix(suffixc);
string left_src_read = lib.get_prefix_split_src_file(lib.get_left_read()) + suffix;
string right_src_read = "";
if (lib.get_paired_end())
right_src_read = lib.get_prefix_split_src_file(lib.get_right_read()) + suffix;
ifstream left_file(left_src_read.c_str());
ifstream right_file;
if (lib.get_paired_end()){
right_file.open(right_src_read.c_str(), ios_base::in);
}
string left_header = "";
string right_header = "";
string left_seq = "";
string right_seq = "";
string left_qual = "";
string right_qual = "";
string plus;
ofstream split_read_fasta_file;
ofstream split_read_fastq_file;
split_read_fasta_file.open(lib.get_split_file_name(file_part, FORMAT_FASTA).c_str(), ios_base::out);
if (lib.get_format() == FORMAT_FASTQ)
split_read_fastq_file.open(lib.get_split_file_name(file_part, FORMAT_FASTQ).c_str(), ios_base::out);
while (getline(left_file, left_header)) {
// get read data point, which includes 4 lines
string lead_chr = (lib.get_format() == FORMAT_FASTQ)? "@" : ">";
if (left_header.substr(0,1) == lead_chr){
//save left-end reads
getline(left_file, left_seq);
if (lib.get_format() == FORMAT_FASTQ) {
getline(left_file, plus);
getline(left_file, left_qual);
}
if (lib.get_paired_end()){
//save right-end reads
while (getline(right_file, right_header))
if (right_header.substr(0,1) == lead_chr)
break;
getline(right_file, right_seq);
if (lib.get_format() == FORMAT_FASTQ) {
getline(right_file, plus);
getline(right_file, right_qual);
}
}
if (lib.get_format() == FORMAT_FASTQ) {
split_read_fastq_file << left_header << endl
<< left_seq << endl
<< "+" << endl
<< left_qual << endl;
}
split_read_fasta_file << ">" << left_header.substr(1) << endl << left_seq << endl;
if (lib.get_paired_end()){
if (lib.get_format() == FORMAT_FASTQ) {
split_read_fastq_file << right_header << endl
<< right_seq << endl
<< "+" << endl
<< right_qual << endl;
}
split_read_fasta_file << ">" << right_header.substr(1) << endl << right_seq << endl;
}
}
}
split_read_fasta_file.close();
if (lib.get_format() == FORMAT_FASTQ)
split_read_fastq_file.close();
left_file.close();
if (lib.get_paired_end())
right_file.close();
string cmd = "rm " + left_src_read + " " + right_src_read;
run_shell_command(cmd);
}
