int generate_snp_sites(char filename[],int output_multi_fasta_file, int output_vcf_file, int output_phylip_file, char output_filename[])
{
size_t length_of_genome;
char * reference_sequence;
int number_of_snps;
int * snp_locations;
int number_of_samples;
int i;
length_of_genome = genome_length(filename);
reference_sequence = (char *) calloc((length_of_genome +1),sizeof(char));
build_reference_sequence(reference_sequence,filename);
number_of_snps = detect_snps(reference_sequence, filename, length_of_genome);
snp_locations = (int *) calloc((number_of_snps+1),sizeof(int));
build_snp_locations(snp_locations, reference_sequence);
free(reference_sequence);
number_of_samples = number_of_sequences_in_file(filename);
// Find out the names of the sequences
char* sequence_names[number_of_samples];
sequence_names[number_of_samples-1] = '\0';
for(i = 0; i < number_of_samples; i++)
{
sequence_names[i] = calloc(MAX_SAMPLE_NAME_SIZE,sizeof(char));
}
get_sample_names_for_header(filename, sequence_names, number_of_samples);
char* bases_for_snps[number_of_snps];
for(i = 0; i < number_of_snps; i++)
{
bases_for_snps[i] = calloc(number_of_samples+1 ,sizeof(char));
}
get_bases_for_each_snp(filename, snp_locations, bases_for_snps, length_of_genome, number_of_snps);
char output_filename_base[FILENAME_MAX];
char filename_without_directory[FILENAME_MAX];
strip_directory_from_filename(filename, filename_without_directory);
strncpy(output_filename_base, filename_without_directory, FILENAME_MAX);
if(output_filename != NULL && *output_filename != '\0')
{
strncpy(output_filename_base, output_filename, FILENAME_MAX);
}
if(output_vcf_file)
{
char vcf_output_filename[FILENAME_MAX];
strncpy(vcf_output_filename, output_filename_base, FILENAME_MAX);
if((output_vcf_file + output_phylip_file + output_multi_fasta_file) > 1 || (output_filename == NULL || *output_filename == '\0') )
{
strcat(vcf_output_filename, ".vcf");
}
create_vcf_file(vcf_output_filename, snp_locations, number_of_snps, bases_for_snps, sequence_names, number_of_samples, length_of_genome);
}
if(output_phylip_file)
{
char phylip_output_filename[FILENAME_MAX];
strncpy(phylip_output_filename, output_filename_base, FILENAME_MAX);
if((output_vcf_file + output_phylip_file + output_multi_fasta_file) > 1 || (output_filename == NULL || *output_filename == '\0') )
{
strcat(phylip_output_filename, ".phylip");
}
create_phylib_of_snp_sites(phylip_output_filename, number_of_snps, bases_for_snps, sequence_names, number_of_samples);
}
if((output_multi_fasta_file) || (output_vcf_file ==0 && output_phylip_file == 0 && output_multi_fasta_file == 0))
{
char multi_fasta_output_filename[FILENAME_MAX];
strncpy(multi_fasta_output_filename, output_filename_base, FILENAME_MAX);
if((output_vcf_file + output_phylip_file + output_multi_fasta_file) > 1 || (output_filename == NULL || *output_filename == '\0') )
{
strcat(multi_fasta_output_filename, ".snp_sites.aln");
}
create_fasta_of_snp_sites(multi_fasta_output_filename, number_of_snps, bases_for_snps, sequence_names, number_of_samples);
}
// free memory
free(snp_locations);
for(i = 0; i < number_of_samples; i++)
{
free(sequence_names[i]);
}
for(i = 0; i < number_of_snps; i++)
{
free(bases_for_snps[i]);
}
return 1;
}
static int generate_snp_sites_generic(char filename[],
int output_multi_fasta_file,
int output_vcf_file,
int output_phylip_file,
char output_filename[],
int output_reference, int pure_mode, int output_monomorphic)
{
int i;
detect_snps(filename, pure_mode, output_monomorphic);
bases_for_snps = calloc(get_number_of_snps()+1, sizeof(char*));
for(i = 0; i < get_number_of_snps(); i++)
{
bases_for_snps[i] = calloc(get_number_of_samples()+1, sizeof(char));
}
get_bases_for_each_snp(filename, bases_for_snps);
char output_filename_base[FILENAME_MAX];
char filename_without_directory[FILENAME_MAX];
strip_directory_from_filename(filename, filename_without_directory);
strncpy(output_filename_base, filename_without_directory, FILENAME_MAX);
if(output_filename != NULL && *output_filename != '\0')
{
strncpy(output_filename_base, output_filename, FILENAME_MAX);
}
if(output_vcf_file)
{
char vcf_output_filename[FILENAME_MAX];
strncpy(vcf_output_filename, output_filename_base, FILENAME_MAX);
if((output_vcf_file + output_phylip_file + output_multi_fasta_file) > 1 || (output_filename == NULL || *output_filename == '\0') )
{
strcat(vcf_output_filename, ".vcf");
}
create_vcf_file(vcf_output_filename, get_snp_locations(), get_number_of_snps(), bases_for_snps, get_sequence_names(), get_number_of_samples(), get_length_of_genome(), get_pseudo_reference_sequence());
}
if(output_phylip_file)
{
char phylip_output_filename[FILENAME_MAX];
strncpy(phylip_output_filename, output_filename_base, FILENAME_MAX);
if((output_vcf_file + output_phylip_file + output_multi_fasta_file) > 1 || (output_filename == NULL || *output_filename == '\0') )
{
strcat(phylip_output_filename, ".phylip");
}
create_phylib_of_snp_sites(phylip_output_filename, get_number_of_snps(), bases_for_snps, get_sequence_names(), get_number_of_samples(), output_reference, get_pseudo_reference_sequence(),get_snp_locations());
}
if((output_multi_fasta_file) || (output_vcf_file ==0 && output_phylip_file == 0 && output_multi_fasta_file == 0))
{
char multi_fasta_output_filename[FILENAME_MAX];
strncpy(multi_fasta_output_filename, output_filename_base, FILENAME_MAX);
if((output_vcf_file + output_phylip_file + output_multi_fasta_file) > 1 || (output_filename == NULL || *output_filename == '\0') )
{
strcat(multi_fasta_output_filename, ".snp_sites.aln");
}
create_fasta_of_snp_sites(multi_fasta_output_filename, get_number_of_snps(), bases_for_snps, get_sequence_names(), get_number_of_samples(), output_reference, get_pseudo_reference_sequence(),get_snp_locations());
}
// free memory
free(get_snp_locations());
for(i = 0; i < get_number_of_samples(); i++)
{
// free(get_sequence_names().[i]);
}
for(i = 0; i < get_number_of_snps(); i++)
{
free(bases_for_snps[i]);
}
free(get_pseudo_reference_sequence());
return 1;
}
void extract_sequences(char vcf_filename[], char tree_filename[],char multi_fasta_filename[],int min_snps, char original_multi_fasta_filename[])
{
FILE *vcf_file_pointer;
vcf_file_pointer=fopen(vcf_filename, "r");
newick_node* root_node;
int number_of_snps;
int number_of_columns;
int i;
int length_of_original_genome;
length_of_original_genome = genome_length(original_multi_fasta_filename);
number_of_columns = get_number_of_columns_from_file(vcf_file_pointer);
char* column_names[number_of_columns];
for(i = 0; i < number_of_columns; i++)
{
column_names[i] = calloc(MAX_SAMPLE_NAME_SIZE,sizeof(char));
}
get_column_names(vcf_file_pointer, column_names, number_of_columns);
number_of_snps = number_of_snps_in_phylip();
int snp_locations[number_of_snps];
get_integers_from_column_in_vcf(vcf_file_pointer, snp_locations, number_of_snps, column_number_for_column_name(column_names, "POS", number_of_columns));
root_node = build_newick_tree(tree_filename, vcf_file_pointer,snp_locations, number_of_snps, column_names, number_of_columns, length_of_original_genome,min_snps);
fclose(vcf_file_pointer);
int filtered_snp_locations[number_of_snps];
int number_of_filtered_snps;
int number_of_samples = number_of_samples_from_parse_phylip();
char * sample_names[number_of_samples];
get_sample_names_from_parse_phylip(sample_names);
char * reference_sequence_bases;
reference_sequence_bases = (char *) calloc((number_of_snps+1),sizeof(char));
get_sequence_for_sample_name(reference_sequence_bases, sample_names[0]);
int internal_nodes[number_of_samples];
int a = 0;
for(a =0; a < number_of_samples; a++)
{
internal_nodes[a] = get_internal_node(a);
}
number_of_filtered_snps = refilter_existing_snps(reference_sequence_bases, number_of_snps, snp_locations, filtered_snp_locations,internal_nodes);
char * filtered_bases_for_snps[number_of_filtered_snps];
filter_sequence_bases_and_rotate(reference_sequence_bases, filtered_bases_for_snps, number_of_filtered_snps);
create_phylip_of_snp_sites(tree_filename, number_of_filtered_snps, filtered_bases_for_snps, sample_names, number_of_samples,internal_nodes);
create_vcf_file(tree_filename, filtered_snp_locations, number_of_filtered_snps, filtered_bases_for_snps, sample_names, number_of_samples,internal_nodes,0);
create_fasta_of_snp_sites(tree_filename, number_of_filtered_snps, filtered_bases_for_snps, sample_names, number_of_samples,internal_nodes);
// Create an new tree with updated distances
scale_branch_distances(root_node, number_of_filtered_snps);
FILE *output_tree_pointer;
output_tree_pointer=fopen(tree_filename, "w");
print_tree(root_node,output_tree_pointer);
fprintf(output_tree_pointer,";");
fflush(output_tree_pointer);
fclose(output_tree_pointer);
// Theres a seg fault in here
for(i = 0; i < number_of_columns; i++)
{
free(column_names[i] );
}
for(i=0; i<number_of_samples; i++ )
{
free(sample_names[i]);
}
for(i=0; i<number_of_filtered_snps; i++ )
{
free(filtered_bases_for_snps[i]);
}
cleanup_node_memory(root_node);
seqFreeAll();
free(reference_sequence_bases);
}