def format_protein_db(self, input_file_path, output_file_path):
progress.new('Formatting raw files')
progress.update('Decompressing protein sequences')
# poor man's uncompress
temp_fasta_path = filesnpaths.get_temp_file_path()
with open(temp_fasta_path, 'wb') as f_out, gzip.open(input_file_path, 'rb') as f_in:
f_out.write(f_in.read())
progress.end()
if utils.is_program_exists('diamond', dont_raise=True):
output_dir = J(self.COG_data_dir, 'DB_DIAMOND')
if os.path.exists(output_dir):
shutil.rmtree(output_dir)
os.mkdir(output_dir)
output_db_path = J(output_dir, 'COG')
log_file_path = J(output_dir, 'log.txt')
self.run.info('Diamond log', log_file_path)
diamond = Diamond(temp_fasta_path)
diamond.num_threads = self.num_threads
diamond.run.log_file_path = log_file_path
diamond.target_db_path = output_db_path
diamond.makedb()
else:
self.run.warning("Diamond does not seem to be installed on this system, so anvi'o is not going to\
generate a search database for it. Remember this when/if things go South.")
if utils.is_program_exists('makeblastdb', dont_raise=True) and utils.is_program_exists('blastp', dont_raise=True):
output_dir = J(self.COG_data_dir, 'DB_BLAST')
if os.path.exists(output_dir):
shutil.rmtree(output_dir)
os.mkdir(output_dir)
output_db_path = J(output_dir, 'COG')
log_file_path = J(output_dir, 'log.txt')
self.run.info('BLAST log', log_file_path)
blast = BLAST(temp_fasta_path)
blast.target_db_path = output_db_path
blast.run.log_file_path = log_file_path
blast.num_threads = self.num_threads
blast.makedb()
else:
self.run.warning("BLAST tools do not seem to be installed on this system, so anvi'o is not going to\
generate a search database for them to be used. Keep this in mind for later.")
os.remove(temp_fasta_path)
def search_with_blastp(self, aa_sequences_file_path):
blast = BLAST(aa_sequences_file_path, run=self.run, progress=self.progress, num_threads=self.num_threads)
blast.target_db_path = self.available_db_search_program_targets['blastp']
self.run.log_file_path = self.log_file_path or J(self.temp_dir_path, 'log.txt')
blast.search_output_path = J(self.temp_dir_path, 'blast-search-results.txt')
blast.max_target_seqs = 1
blast.blastp()
return blast.search_output_path
def run_blast(self, unique_proteins_fasta_path, unique_proteins_names_dict):
self.run.warning("You elected to use NCBI's blastp for protein search. Running blastp will be significantly\
slower than DIAMOND (although, anvi'o developers are convinced that you *are*\
doing the right thing, so, kudos to you).")
blast = BLAST(unique_proteins_fasta_path, run=self.run, progress=self.progress,
num_threads=self.num_threads, overwrite_output_destinations=self.overwrite_output_destinations)
blast.names_dict = unique_proteins_names_dict
blast.log_file_path = self.log_file_path
blast.target_db_path = self.get_output_file_path(filesnpaths.get_name_from_file_path(unique_proteins_fasta_path))
blast.search_output_path = self.get_output_file_path('blast-search-results.txt')
return blast.get_blastall_results()
def run_blast(self, unique_proteins_fasta_path, unique_proteins_names_dict):
self.run.warning("You elected to use NCBI's blastp for protein search. Running blastp will be significantly\
slower than DIAMOND (although, anvi'o developers are convinced that you *are*\
doing the right thing, so, kudos to you).")
blast = BLAST(unique_proteins_fasta_path, run=self.run, progress=self.progress,
num_threads=self.num_threads, overwrite_output_destinations=self.overwrite_output_destinations)
blast.names_dict = unique_proteins_names_dict
blast.log_file_path = self.log_file_path
blast.target_db_path = self.get_output_file_path(filesnpaths.get_name_from_file_path(unique_proteins_fasta_path))
blast.search_output_path = self.get_output_file_path('blast-search-results.txt')
return blast.get_blastall_results()
