WT_Boolean WT_Alignment::operator== (WT_Alignment const &
alignment) const
{
if(get_alignment_description() == alignment.get_alignment_description())
return WD_True;
return WD_False;
}
/****************************************************************************
* Remove from an alignment any sequence whose ID is not in a given list.
*
* N.B. It is NOT an error for the given list to contain sequence IDs that
* are not in the alignment.
****************************************************************************/
ALIGNMENT_T* remove_alignment_seqs
(STRING_LIST_T* seqs_to_keep,
ALIGNMENT_T* alignment)
{
// Extract the names of the sequences in the alignment.
STRING_LIST_T* alignment_species = get_species_names(alignment);
int num_species = get_num_strings(alignment_species);
// Count how many sequences will be in the new alignment.
int i_species;
int num_final = 0;
for (i_species = 0; i_species < num_species; i_species++) {
char* this_species = get_nth_string(i_species, alignment_species);
if (have_string(this_species, seqs_to_keep)) {
num_final++;
} else {
if (verbosity >= NORMAL_VERBOSE) {
fprintf(stderr, "Removing %s from alignment.\n", this_species);
}
}
}
// Allocate space for the new sequences.
SEQ_T** new_sequences = (SEQ_T**)mm_malloc(num_final * sizeof(SEQ_T*));
// Copy the sequences.
int final_index = 0;
num_species = get_num_strings(seqs_to_keep);
for (i_species = 0; i_species < num_species; i_species++) {
char* this_species = get_nth_string(i_species, seqs_to_keep);
// If the requested ID is in the alignment, then copy over the sequence.
if (have_string(this_species, alignment_species)) {
SEQ_T* this_seq
= get_alignment_sequence_by_name(this_species, alignment);
new_sequences[final_index] = copy_sequence(this_seq);
final_index++;
}
}
// Allocate and return the new alignment.
char *consensus = NULL;
copy_string(&consensus, get_consensus_string(alignment));
ALIGNMENT_T* new_alignment
= allocate_alignment(get_alignment_name(alignment),
get_alignment_description(alignment),
num_final,
new_sequences,
consensus);
return(new_alignment);
}
/****************************************************************************
* Extract a small alignment out of the middle of a larger alignment.
****************************************************************************/
ALIGNMENT_T* extract_subalignment
(int start,
int width,
ALIGNMENT_T* alignment)
{
int num_sequences = get_num_aligned_sequences(alignment);
SEQ_T** sequences = get_alignment_sequences(alignment);
SEQ_T** subsequences = (SEQ_T**)mm_malloc(num_sequences * sizeof(SEQ_T*));
// Extract the specified columns into a new list of sequences.
int i_seq = 0;
char* subsequence = mm_malloc((width + 1) * sizeof(char));
for (i_seq = 0; i_seq < num_sequences; i_seq++) {
SEQ_T* this_seq = sequences[i_seq];
char* raw_seq = get_raw_sequence(this_seq);
strncpy(subsequence, raw_seq + start, width);
subsequence[width] = '\0';
subsequences[i_seq] =
allocate_seq(get_seq_name(this_seq),
get_seq_description(this_seq),
get_seq_offset(this_seq),
subsequence);
}
// Extract the consensus string in the specified columns.
char* consensus = get_consensus_string(alignment);
char* subconsensus = mm_malloc(sizeof(char) * (width + 1));
strncpy(subconsensus, consensus + start, width);
subconsensus[width] = '\0';
// Allocate and return the new alignment.
ALIGNMENT_T* subalignment
= allocate_alignment(get_alignment_name(alignment),
get_alignment_description(alignment),
num_sequences,
subsequences,
subconsensus);
// Free local dynamic memory.
for (i_seq = 0; i_seq < num_sequences; i_seq++) {
free_seq(subsequences[i_seq]);
}
myfree(subsequences);
myfree(subsequence);
return(subalignment);
}
WT_Result WT_Alignment::serialize(WT_File & file) const
{
WD_CHECK (file.dump_delayed_drawable());
if (file.heuristics().allow_binary_data())
{
WD_CHECK (file.write((WT_Byte) '{'));
WD_CHECK (file.write((WT_Integer32) (sizeof(WT_Unsigned_Integer16) + // for the opcode
sizeof(WT_Unsigned_Integer16) + // orientation description
sizeof(WT_Byte) // The closing "}"
)));
WD_CHECK (file.write((WT_Unsigned_Integer16) WD_EXBO_ALIGNMENT));
WD_CHECK (file.write((WT_Unsigned_Integer16)
get_alignment_description()));
WD_CHECK (file.write((WT_Byte) '}'));
}
else {
//Extended ASCII format
WD_CHECK (file.write_geom_tab_level());
WD_CHECK (file.write("(Alignment "));
if(get_alignment_description() == WT_Alignment::Align_Center) {
WD_CHECK (file.write_quoted_string("Align_Center "));
}
else if (get_alignment_description() == WT_Alignment::Align_Title_Block) {
WD_CHECK (file.write_quoted_string("Align_Title_Block ", WD_True));
}
else if (get_alignment_description() == WT_Alignment::Align_Top) {
WD_CHECK (file.write_quoted_string("Align_Top ", WD_True));
}
else if (get_alignment_description() == WT_Alignment::Align_Bottom) {
WD_CHECK (file.write_quoted_string("Align_Bottom ", WD_True));
}
else if (get_alignment_description() == WT_Alignment::Align_Left) {
WD_CHECK (file.write_quoted_string("Align_Left ", WD_True));
}
else if (get_alignment_description() == WT_Alignment::Align_Right) {
WD_CHECK (file.write_quoted_string("Align_Right ", WD_True));
}
else if (get_alignment_description() == WT_Alignment::Align_Top_Left) {
WD_CHECK (file.write_quoted_string("Align_Top_Left ", WD_True));
}
else if (get_alignment_description() == WT_Alignment::Align_Top_Right) {
WD_CHECK (file.write_quoted_string("Align_Top_Right ", WD_True));
}
else if (get_alignment_description() == WT_Alignment::Align_Bottom_Left) {
WD_CHECK (file.write_quoted_string("Align_Bottom_Left ", WD_True));
}
else if (get_alignment_description() == WT_Alignment::Align_Bottom_Right) {
WD_CHECK (file.write_quoted_string("Align_Bottom_Right", WD_True));
}
else if (get_alignment_description() == WT_Alignment::Align_None) {
WD_CHECK (file.write_quoted_string("Align_None ", WD_True));
}
else
return WT_Result::Internal_Error;
WD_CHECK (file.write(")"));
}
return WT_Result::Success;
}
/****************************************************************************
* Remove from the alignment all columns that contain gaps for the
* specified species.
****************************************************************************/
ALIGNMENT_T* remove_alignment_gaps
(char* species,
ALIGNMENT_T* alignment)
{
// Locate this species in the alignment.
int species_index = get_index_in_string_list(species,
get_species_names(alignment));
if (species_index == -1) {
die("Can't find %s in alignment.\n", species);
}
SEQ_T* this_seq = get_alignment_sequence(species_index, alignment);
// Get the dimensions of the original matrix.
int num_sequences = get_num_aligned_sequences(alignment);
int alignment_length = get_alignment_length(alignment);
// Allocate memory for raw sequences that will constitute the new alignment.
char** raw_sequences = (char**)mm_malloc(sizeof(char*) * num_sequences);
int i_seq = 0;
for (i_seq = 0; i_seq < num_sequences; i_seq++) {
raw_sequences[i_seq]
= (char*)mm_calloc(alignment_length + 1, sizeof(char*));
}
char* consensus = get_consensus_string(alignment);
char* new_consensus
= (char*)mm_calloc(alignment_length + 1, sizeof(char*));
// Iterate over all columns.
int i_column;
int i_raw = 0;
for (i_column = 0; i_column < alignment_length; i_column++) {
// Is there a gap?
char this_char = get_seq_char(i_column, this_seq);
if ((this_char != '-') && (this_char != '.')) {
// If no gap, then copy over this column.
for (i_seq = 0; i_seq < num_sequences; i_seq++) {
SEQ_T* this_sequence = get_alignment_sequence(i_seq, alignment);
char this_char = get_seq_char(i_column, this_sequence);
raw_sequences[i_seq][i_raw] = this_char;
}
new_consensus[i_raw] = consensus[i_column];
i_raw++;
}
}
// Create new sequence objects.
SEQ_T** new_sequences = (SEQ_T**)mm_malloc(num_sequences * sizeof(SEQ_T*));
for (i_seq = 0; i_seq < num_sequences; i_seq++) {
SEQ_T* this_sequence = get_alignment_sequence(i_seq, alignment);
new_sequences[i_seq] = allocate_seq(get_seq_name(this_sequence),
get_seq_description(this_sequence),
get_seq_offset(this_sequence),
raw_sequences[i_seq]);
}
// Allocate and return the new alignment.
ALIGNMENT_T* new_alignment
= allocate_alignment(get_alignment_name(alignment),
get_alignment_description(alignment),
num_sequences,
new_sequences,
new_consensus);
// Free local dynamic memory.
for (i_seq = 0; i_seq < num_sequences; i_seq++) {
myfree(raw_sequences[i_seq]);
free_seq(new_sequences[i_seq]);
}
myfree(raw_sequences);
myfree(new_sequences);
myfree(new_consensus);
return(new_alignment);
}
