void test_read_fasta_file(struct test *t) {
t_set_msg(t, "Testing reading a fasta File...");
struct genome_sequence *sequence_table = NULL;
char filename[30] = "test/data/test.fasta";
int err = read_fasta_file(&sequence_table, filename);
t_assert_msg(t, err == E_SUCCESS, "Parsing failed");
t_log(t, "Error: %d\n", err);
struct genome_sequence *s = NULL;
struct genome_sequence *tmp = NULL;
HASH_ITER(hh, sequence_table, s, tmp) {
s->data[s->n] = 0;
t_log(t, "%s %s \n\n", s->chrom, s->data);
}
void ReadGenome(const vector<string>& chrom_files, Genome& genome) {
fprintf(stderr, "[READING CHROMOSOMES]\n");
vector<string> chrom_names;
vector<string> chrom_seqs;
uint32_t all_chroms_len = 0;
for (uint32_t i = 0; i < chrom_files.size(); ++i) {
vector<string> tmp_chrom_names;
vector<string> tmp_chrom_seqs;
/* read chromosome name and seqeunce from the chromosome file */
read_fasta_file(chrom_files[i].c_str(), tmp_chrom_names, tmp_chrom_seqs);
for (uint32_t j = 0; j < tmp_chrom_seqs.size(); ++j) {
chrom_names.push_back(tmp_chrom_names[j]);
chrom_seqs.push_back(tmp_chrom_seqs[j]);
all_chroms_len += tmp_chrom_seqs[j].size();
}
}
/* copy chroms sequences to genome */
genome.num_of_chroms = chrom_seqs.size();
genome.length_of_genome = all_chroms_len;
fprintf(stderr, "[THERE ARE %u CHROMOSOMES IN THE GENOME]\n",
genome.num_of_chroms);
fprintf(stderr, "[THE TOTAL LENGTH OF ALL CHROMOSOMES IS %u]\n",
all_chroms_len);
genome.name.resize(genome.num_of_chroms);
genome.length.resize(genome.num_of_chroms);
/* The last element of genome.start-index is the length of all chroms */
genome.start_index.resize(genome.num_of_chroms + 1);
genome.strand = '+';
genome.sequence.resize(all_chroms_len);
uint32_t k = 0;
for (uint32_t i = 0; i < genome.num_of_chroms; ++i) {
genome.name[i] = chrom_names[i];
genome.length[i] = chrom_seqs[i].size();
genome.start_index[i] = k;
for (uint32_t j = 0; j < chrom_seqs[i].size(); ++j) {
genome.sequence[k] = toupper(chrom_seqs[i][j]);
genome.sequence[k] = toACGT(genome.sequence[k]);
k++;
}
}
genome.start_index[genome.num_of_chroms] = k;
}
// unit test: read fasta file and print all entries
int main(int argc, char * argv[])
{
if (argc < 2)
{
puts ("ERROR: no FASTA-file specified!");
return 1;
}
multifasta *mf = read_fasta_file (argv[1]);
if (mf == NULL)
return 1;
int i;
for (i = 0; i < mf->length; i++)
{
printf("Entry %d \n", i);
printf("Header: %s \n", mf->entries[i]->header);
printf("Sequenz: %s \n", mf->entries[i]->sequence);
}
multifasta_delete (mf);
return 0;
}
int main ( int argc, char ** argv)
{
struct TSwitch sw;
char * out_file;
unsigned int MAXnumgaps; //input argument
double gap_open_pen; //input argument
double gap_extend_pen; //input argument
unsigned int scoring_matrix; //input argument
unsigned int MAXgap;
unsigned int num_mat; //input argument + 1
double MAXscore; //to be computed
unsigned int MINnumgaps; //to be computed
unsigned int istart; //where to start backtracing
unsigned int jstart; //where to start backtracing
unsigned int iend; //where to end backtracing
unsigned int jend; //where to end backtracing
unsigned int * gaps_pos; //position of the gap(s)
unsigned int * gaps_len; //len of the gap(s)
unsigned int * where; //where is the gap(s): text [1] or pattern [2]
struct TSeq * t; //text t
unsigned int n; //length of text
struct TSeq * p; //pattern p
unsigned int m; //length of pattern
unsigned int L; //local alignment
double *** G; //dynamic programming matrix
int *** H; //backtracing matrix
unsigned int swap; //swap the text and the pattern in case m < n
unsigned int i, j;
/* checks the arguments */
i = decode_switches ( argc, argv, &sw );
if ( i < 5 || ! sw . seq_a || ! sw . seq_b )
{
usage ();
return ( 1 );
}
else
{
gap_open_pen = - sw . gap_open_pen; //the penalties should have a negative value
gap_extend_pen = - sw . gap_extend_pen;
out_file = sw . out_file;
L = sw . L;
if ( ! strcmp ( "EDNAFULL", sw . matrix ) ) scoring_matrix = 0;
else if ( ! strcmp ( "EBLOSUM62", sw . matrix ) ) scoring_matrix = 1;
else
{
fprintf ( stderr, "Error: scoring matrix argument should be `EDNAFULL' for nucleotide sequences or `EBLOSUM62' for protein sequences!!!\n" );
return ( 1 );
}
if ( L > 1 )
{
fprintf (stderr, "Error: wrong argument with L: should be either 0 (semi-global) or 1 (local)\n" );
return ( 1 );
}
}
/* reads the text */
t = read_fasta_file ( sw . seq_a );
if ( ! t )
{
fprintf (stderr, "Error: cannot read file %s!!!\n", sw . seq_a );
return ( 1 );
}
if ( ! t -> header ) t -> header = strdup ( "Seq A" );
/* reads the pattern */
p = read_fasta_file ( sw . seq_b );
if ( ! p )
{
fprintf( stderr, "Error: cannot read file %s!!!\n", sw . seq_b );
return ( 1 );
}
if ( ! p -> header ) p -> header = strdup ( "Seq B" );
/* calculate text's and pattern's length */
n = strlen ( t -> data );
m = strlen ( p -> data );
/* checks the lengths of text and pattern and swaps if needed */
if ( m > n )
{
swap_txt_pat ( &t, &n, &p, &m );
swap = 1;
}
else
swap = 0;
/* checks the max num of gaps: MAXnumgaps < n */
MAXnumgaps = ( sw . max_num_gaps <= -1 ) ? 2 : sw . max_num_gaps;
if( MAXnumgaps >= n )
{
fprintf ( stderr, "Error: the max gap length should be less than the length of the text!!!\n" );
return ( 1 );
}
if ( L == 1 ) num_mat = MAXnumgaps + 1;
else num_mat = MAXnumgaps;
MAXgap = ( sw . max_gap <= -1 ) ? n - 1 : sw . max_gap;
if( MAXgap >= n )
{
fprintf ( stderr, "Error: the max gap length should be less than the length of the text!!!\n" );
return ( 1 );
}
/* 3d dynamic memory allocation for matrices G and H*/
if( ( G = ( double *** ) malloc ( ( num_mat ) * sizeof( double ** ) ) ) == NULL )
{
fprintf( stderr, "G could not be allocated\n" );
return 0;
}
for ( i = 0; i < num_mat; i ++ )
{
if( ( G[i] = ( double ** ) malloc ( ( n + 1 ) * sizeof( double * ) ) ) == NULL )
{
fprintf( stderr, "G could not be allocated\n" );
return 0;
}
if( ( G[i][0] = ( double * ) calloc ( ( n + 1 ) * ( m + 1 ), sizeof( double ) ) ) == NULL )
{
fprintf( stderr, "G could not be allocated\n" );
return 0;
}
for ( j = 1; j < n + 1; j++ )
G[i][j] = ( void * ) G[i][0] + j * ( m + 1 ) * sizeof( double );
}
if( ( H = ( int *** ) malloc ( ( num_mat ) * sizeof( int ** ) ) ) == NULL )
{
fprintf( stderr, "H could not be allocated\n" );
return 0;
}
for ( i = 0; i < num_mat; i ++ )
{
if( ( H[i] = ( int ** ) malloc ( ( n + 1 ) * sizeof( int * ) ) ) == NULL )
{
fprintf( stderr, "H could not be allocated\n" );
return 0;
}
if( ( H[i][0] = ( int * ) calloc ( ( n + 1 ) * ( m + 1 ), sizeof( int ) ) ) == NULL )
{
fprintf( stderr, "H could not be allocated\n" );
return 0;
}
for ( j = 1; j < n + 1; j++ )
H[i][j] = ( void* ) H[i][0] + j * ( m + 1 ) * sizeof( int );
}
/* dynamic programming algorithm */
if ( L == 0 )
{
if ( MAXgap == n - 1 )
{
if ( ! ( dp_algorithm( G, MAXnumgaps, H, t -> data, n, p -> data, m, scoring_matrix, gap_open_pen, gap_extend_pen ) ) )
{
fprintf ( stderr, "Error: dp_algorithm() failed!!!\n" );
return ( 1 );
}
}
else
{
if ( ! ( dp_algorithm_pruned( G, MAXnumgaps, H, t -> data, n, p -> data, m, scoring_matrix, gap_open_pen, gap_extend_pen, MAXgap ) ) )
{
fprintf ( stderr, "Error: dp_algorithm_pruned() failed!!!\n" );
return ( 1 );
}
}
}
else
{
if ( MAXgap == n - 1 )
{
if ( ! ( dp_algorithm_lcl( G, MAXnumgaps, H, t -> data, n, p -> data, m, scoring_matrix, gap_open_pen, gap_extend_pen ) ) )
{
fprintf ( stderr, "Error: dp_algorithm() failed!!!\n" );
return ( 1 );
}
}
else
{
fprintf ( stderr, "Error: option `- m' is only for global alignment!!!\n" );
return ( 1 );
}
}
MINnumgaps = 0; //to be computed
/* finds the optimal alignment based on the matrices scores */
if ( L == 0 ) opt_solution ( G, MAXnumgaps, n, m, &MAXscore, &istart, &MINnumgaps );
if ( L == 1 ) opt_solution_lcl ( G, MAXnumgaps, n, m, &MAXscore, &istart, &jstart, &MINnumgaps );
if( ( gaps_pos = ( unsigned int * ) calloc ( MINnumgaps, sizeof( unsigned int ) ) ) == NULL )
{
fprintf( stderr, "gaps_pos could not be allocated\n" );
return 0;
}
if( ( gaps_len = ( unsigned int * ) calloc ( MINnumgaps, sizeof( unsigned int ) ) ) == NULL )
{
fprintf( stderr, "gaps_pos could not be allocated\n" );
return 0;
}
if( ( where = ( unsigned int * ) calloc ( MINnumgaps, sizeof( unsigned int ) ) ) == NULL )
{
fprintf( stderr, "where could not be allocated\n" );
return 0;
}
/* computes the position of the gap */
if ( L == 0 ) backtracing ( H[MINnumgaps - 1], m, n, istart, gaps_pos, MINnumgaps, gaps_len, where );
if ( L == 1 ) backtracing_lcl ( G[MINnumgaps], m, n, H[MINnumgaps], istart, jstart, gaps_pos, MINnumgaps, gaps_len, where, &iend, &jend );
#if 0
int s;
for ( s = 0; s < num_mat; s++ )
{
for(i = 0; i < n+1; i++) //Matrix G output
{
for(j = 0; j < m+1; j++)
{
fprintf( stderr,"%d\t", (int) G[s][i][j] );
}
fprintf(stderr,"\n");
}
fprintf(stderr,"\n");
#if 1
for(i = 0; i < n+1; i++) //Matrix H output
{
for(j = 0; j < m+1; j++)
{
fprintf( stderr,"%d\t", H[s][i][j]);
}
fprintf(stderr,"\n");
}
fprintf(stderr,"\n");
#endif
}
if ( L == 1 ) fprintf( stderr,"\n[%d,%d]-->[%d,%d]\n", istart, jstart, iend, jend );
#endif
/* outputs the results */
if ( L == 0 )
if ( ! ( results ( out_file, t, n, p, m, MAXscore, gaps_pos, MINnumgaps, gaps_len, where, swap, scoring_matrix, gap_open_pen, gap_extend_pen, L ) ) )
{
fprintf(stderr, "Error: results() failed!!!\n");
return ( 1 );
}
if ( L == 1 )
if ( ! ( results_lcl ( out_file, t, n, p, m, MAXscore, gaps_pos, MINnumgaps, gaps_len, where, istart, iend, jstart, jend, swap, scoring_matrix, gap_open_pen, gap_extend_pen, L ) ) )
{
fprintf(stderr, "Error: results() failed!!!\n");
return ( 1 );
}
for ( i = 0; i < num_mat; i++ )
{
free ( G[i][0] );
free ( G[i] );
free ( H[i][0] );
free ( H[i] );
}
free ( G );
free ( H );
free ( gaps_pos );
free ( gaps_len );
free ( where );
free ( t -> header );
free ( p -> header );
free ( t );
free ( p );
free ( sw . out_file );
free ( sw . matrix );
return ( 0 );
}
