int main
(int argc,
char * argv[])
{
// Parse the command line.
if (argc != 4) {
fprintf(stderr, "Usage: reduce-alignment <start> <width> <alignment>\n");
exit(1);
}
int start_position = atoi(argv[1]);
int width = atoi(argv[2]);
char* alignment_filename = argv[3];
// Read the alignment.
ALIGNMENT_T* big_alignment = read_alignment_from_file(
alignment_filename,
FALSE,
FALSE,
NULL // pointer to ref_seq_index, not used.
);
fprintf(stderr, "Read alignment of %d sequences and %d columns.\n",
get_num_aligned_sequences(big_alignment),
get_alignment_length(big_alignment));
if (start_position + width > get_alignment_length(big_alignment)) {
fprintf(stderr, "Invalid coordinates: %d + %d > %d.\n",
start_position, width, get_alignment_length(big_alignment));
exit(1);
}
// Extract the smaller alignment.
ALIGNMENT_T* small_alignment = extract_subalignment(start_position,
width,
big_alignment);
fprintf(stderr, "Created alignment of %d sequences and %d columns.\n",
get_num_aligned_sequences(small_alignment),
get_alignment_length(small_alignment));
// Print the alignment.
print_clustalw(stdout, FALSE, small_alignment);
// Free locally allocated memory.
free_alignment(big_alignment);
free_alignment(small_alignment);
return(0);
}
/****************************************************************************
* Return a list containing the empirical column frequency distributions
* for all alignments in the input.
*
* Each file in the list of filenames is read and the species list is
* determined. The counts of each occurring column are tallied.
* All files with the same species lists get their counts combined.
*
* The returned list contains one distribution per species list that
* occurs in some alignment.
****************************************************************************/
OBJECT_LIST_T* get_alignment_column_freqs_list
(ALPH_T alph,
STRING_LIST_T* filenames,
BOOLEAN_T remove_allgap_seqs)
{
int file_index;
int num_filenames = get_num_strings(filenames);
ARRAY_T* alignment_column_freqs = NULL;
OBJECT_LIST_T* alignment_column_freqs_list
= new_object_list(equal_string_lists,
(void*)copy_string_list,
free_string_list,
free_array);
// Consider each alignment in turn.
for(file_index = 0; file_index < num_filenames; file_index++) {
char* filename = get_nth_string(file_index, filenames);
if (verbosity >= NORMAL_VERBOSE && !(file_index % 1)) {
fprintf(
stderr,
"Computing column freqs: alignment file number %d of %d total files.\n",
file_index+1, num_filenames
);
}
// Read the alignment
int ref_seq_index = 0;
ALIGNMENT_T* alignment =
read_alignment_from_file(filename, TRUE, remove_allgap_seqs, &ref_seq_index);
STRING_LIST_T* alignment_species = get_species_names(alignment);
// Try to retrieve the counts so far for this list of species.
alignment_column_freqs =
(ARRAY_T*)retrieve_object(
alignment_species,
alignment_column_freqs_list
);
// Found counts for current species list?
if (alignment_column_freqs) {
// Add counts from current alignment.
(void) build_alignment_column_counts(alph, alignment, alignment_column_freqs);
// Note: objects in lists are references, so no need to re-store
// after modification.
}
// Didn't find counts for this species list, so create new array of counts.
else {
alignment_column_freqs = build_alignment_column_counts(alph, alignment, NULL);
store_object(
(void*)alignment_column_freqs,
(void*)alignment_species,
0.0, // Score
alignment_column_freqs_list
);
}
// free space used by alignment
free_alignment(alignment);
} // each filename
fprintf(stderr, "\n");
// Convert counts to frequencies by retrieving each array of counts
// and dividing by the total counts for that list of species.
while (
(alignment_column_freqs = retrieve_next_object(alignment_column_freqs_list)
) != NULL )
{
int i;
int num_freqs = get_array_length(alignment_column_freqs);
double total_counts;
// Get total counts.
for (i=total_counts=0; i<num_freqs; i++) {
total_counts += get_array_item(i, alignment_column_freqs);
}
// Get frequencies.
for (i=0; i<num_freqs; i++) {
double f = get_array_item(i, alignment_column_freqs);
set_array_item(i, f/total_counts, alignment_column_freqs);
#ifdef DEBUG
int asize = alph_size(alph, ALPH_SIZE);
int num_leaves = NINT(log(num_freqs)/log(asize));
char* alignment_col = mm_malloc((num_leaves + 1) * sizeof(char));
unhash_alignment_col(
alph,
i, //col_index
alignment_col,
num_leaves
);
printf("%s %g %g\n", alignment_col, f, f/total_counts);
myfree(alignment_col);
#endif
} // get frequencies
} // while more species lists
return(alignment_column_freqs_list);
} // get_alignment_column_freqs_list
