def predict_genes(self,
output_prefix,
annotation_species_prefix,
genome_fasta,
augustus_species,
output_directory="./",
augustus_strand=None,
augustus_gene_model=None,
augustus_config_dir=None,
augustus_use_softmasking=None,
augustus_other_options="",
augustus_hintsfile=None,
augustus_extrinsicCfgFile=None,
augustus_predict_UTR=None,
augustus_min_intron_len=None,
threads=1,
augustus_dir="",
hmmer_dir="",
blast_dir="",
stop_codons_list=("TGA", "TAA", "TAG"),
genetic_code_table=1):
draft_file_prefix = "%s/raw/%s" % (output_directory, output_prefix)
augustus_splited_input_dir = "%s/splited_input/" % output_directory
augustus_splited_output_dir = "%s/splited_output_dir" % output_directory
output_raw_gff = "%s.raw.gff" % draft_file_prefix
output_gff = "%s.renamed.gff" % draft_file_prefix
augustus_pep = "%s.pep" % draft_file_prefix
AUGUSTUS.path = augustus_dir
AUGUSTUS.threads = threads
HMMER3.path = hmmer_dir
HMMER3.threads = threads
BLASTp.path = blast_dir
BLASTp.threads = threads
print("Annotating genes...")
AUGUSTUS.parallel_predict(
augustus_species,
genome_fasta,
output_raw_gff,
strand=augustus_strand,
gene_model=augustus_gene_model,
output_gff3=True,
other_options=augustus_other_options,
config_dir=augustus_config_dir,
use_softmasking=augustus_use_softmasking,
hints_file=augustus_hintsfile,
split_dir=augustus_splited_input_dir,
splited_output_dir=augustus_splited_output_dir,
extrinsicCfgFile=augustus_extrinsicCfgFile,
predict_UTR=augustus_predict_UTR,
combine_output_to_single_file=True,
min_intron_len=augustus_min_intron_len)
#replace_augustus_ids(augustus_gff, output_prefix, species_prefix=None, number_of_digits_in_id=8):
AUGUSTUS.replace_augustus_ids(output_raw_gff,
draft_file_prefix,
species_prefix=annotation_species_prefix,
number_of_digits_in_id=8)
#extract_transcript_sequences(self, input_gff_file, genomic_fasta_file, output_prefix, coding_only=False)
gffread_file_prefix = "%s.gffread" % draft_file_prefix
gffread_transcripts_file, gffread_cds_file, gffread_pep_file = Gffread.extract_transcript_sequences(
output_gff, genome_fasta, gffread_file_prefix)
gffread_trimmed_cds = ".".join(
gffread_cds_file.split(".")[:-1]) + ".trimmed.cds"
gffread_trimmed_pep = ".".join(
gffread_pep_file.split(".")[:-1]) + ".trimmed.pep"
self.trim_cds_and_remove_terminal_stop_codons(
gffread_cds_file,
gffread_trimmed_cds,
stop_codons_list=stop_codons_list
) # using default stop_codons(from universal genetic_code)/ Note that this will affect mtDNA proteins
inframe_stop_codons_file_prefix = "%s.inframe_stop_codon" % draft_file_prefix
self.translate_sequences_from_file(
gffread_trimmed_cds,
gffread_trimmed_pep,
format="fasta",
id_expression=None,
genetic_code_table=genetic_code_table,
translate_to_stop=False,
prefix_of_file_inframe_stop_codons_seqsin=
inframe_stop_codons_file_prefix) # Universal code !!!
AUGUSTUS.extract_gene_ids_from_output(output_gff,
all_annotated_genes_ids)
AUGUSTUS.extract_CDS_annotations_from_output(output_gff, CDS_gff)
print("Extracting peptides...")
AUGUSTUS.extract_proteins_from_output(
output_gff,
output_pep,
id_prefix="",
evidence_stats_file=output_evidence_stats,
supported_by_hints_file=output_supported_stats)
self.compare_sequences_from_files(output_pep,
"%s.trimmed.pep" % args.output,
"comparison_of_peptides",
format="fasta",
verbose=True)
os.system("awk -F'\\t' 'NR==1 {}; NR > 1 {print $2}' %s > %s" %
(output_supported_stats, output_supported_stats_ids))
print("Annotating domains(Pfam database)...")
HMMER3.parallel_hmmscan(
args.pfam_db,
output_pep,
output_hmmscan,
num_of_seqs_per_scan=None,
split_dir="splited_hmmscan_fasta/",
splited_output_dir="splited_hmmscan_output_dir",
tblout_outfile=None,
domtblout_outfile=output_domtblout,
pfamtblout_outfile=None,
splited_tblout_dir=None,
splited_domtblout_dir="hmmscan_domtblout/")
HMMER3.extract_dom_ids_hits_from_domtblout(
output_domtblout, output_pfam_annotated_dom_ids)
hits_dict = HMMER3.extract_dom_names_hits_from_domtblout(
output_domtblout, output_pfam_annotated_dom_names)
supported_ids = IdSet(hits_dict.keys())
supported_ids.write(output_pfam_supported_transcripts_ids)
remove_transcript_ids_str = "sed -re 's/\.t[0123456789]+//' %s | sort -k 1 | uniq > %s" % (
output_pfam_supported_transcripts_ids,
output_pfam_supported_genes_ids)
os.system(remove_transcript_ids_str)
print("Annotating peptides(Swissprot database)...")
BLASTp.parallel_blastp(output_pep,
args.swissprot_db,
evalue=0.0000001,
output_format=6,
outfile=output_swissprot_blastp_hits,
split_dir="splited_blastp_fasta",
splited_output_dir="splited_blastp_output_dir")
hits_dict = BLASTp.extract_hits_from_tbl_output(
output_swissprot_blastp_hits, output_swissprot_blastp_hits_names)
supported_ids = IdSet(hits_dict.keys())
supported_ids.write(output_swissprot_supported_transcripts_ids)
remove_transcript_ids_str = "sed -re 's/\.t[0123456789]+//' %s | sort -k 1 | uniq > %s" % (
output_swissprot_supported_transcripts_ids,
output_swissprot_supported_genes_ids)
os.system(remove_transcript_ids_str)
"""
bad_antigen_candidates_coordinates = "%s.bad_candidates.coordinates" % args.out_prefix
bad_antigen_candidates_coordinates_sorted = "%s.bad_candidates.sorted.coordinates" % args.out_prefix
BLASTp.threads = args.threads
sequence = list(SeqIO.parse(args.input, format="fasta"))[0]
print("Constructing kmer list...\n")
#print len(sequence.seq)
kmer_dict = get_kmer_dict_as_seq_records(sequence.seq, args.length, args.start, args.end)
kmer_ids = list(kmer_dict.keys())
SeqIO.write(record_by_expression_generator(kmer_dict), kmer_file, format="fasta")
print("Blast of kmers vs species peptides\n")
BLASTp.search(kmer_file, args.species_db, outfile=species_blast_hits,
blast_options=None, evalue=args.species_evalue, output_format=6)
species_grep_string = "grep -v %s %s > %s" % ("|".join(args.protein_ids), species_blast_hits,
species_blast_hits_no_self_hits)
species_awk_string = "awk '{print$1}' %s | uniq > %s" % (species_blast_hits_no_self_hits,
species_blast_hits_no_self_hits_ids)
os.system(species_grep_string)
os.system(species_awk_string)
print("Blast of kmers vs immunogenetic species peptides\n")
BLASTp.search(kmer_file, args.immune_db, outfile=immune_blast_hits,
blast_options=None, evalue=args.immune_evalue, output_format=6)
immune_awk_string = "awk '{print$1}' %s | uniq > %s" % (immune_blast_hits,
immune_blast_hits_ids)
os.system(immune_awk_string)
genetic_code=args.genetic_code,
analyze_only_top_strand=args.analyze_only_top_strand,
minimum_protein_length=args.min_prot_len)
if args.pfam_database:
HMMER3.parallel_hmmscan(args.pfam_database,
pep_from_longest_orfs,
hmmscan_vs_pfam_output,
split_dir=hmmscan_splited_fasta_dir,
splited_domtblout_dir=splited_domtblout_dir,
domtblout_outfile=domtblout_outfile,
dont_output_alignments=True)
if args.blast_database:
BLASTp.parallel_blastp(pep_from_longest_orfs,
args.blast_database,
outfile=blastp_outfile,
evalue=0.00001,
output_format=6,
blast_options=" -max_target_seqs 1",
combine_output_to_single_file=True,
split_dir=blastp_split_dir,
splited_output_dir=blastp_splited_output_dir)
TransDecoder.predict_pep(
args.input,
pfam_hits=domtblout_outfile,
blastp_hits=blastp_outfile,
minimum_orf_length_if_no_other_evidence=args.
min_orf_len_if_no_other_evidence,
file_with_orfs_for_training=args.file_with_orfs_for_training,
number_of_top_orfs_for_training=args.number_of_top_orfs_for_training)
output_pfam_annotated_dom_ids)
hits_dict = HMMER3.extract_dom_names_hits_from_domtblout(
output_domtblout, output_pfam_annotated_dom_names)
supported_ids = IdSet(hits_dict.keys())
supported_ids.write(output_pfam_supported_transcripts_ids)
remove_transcript_ids_str = "sed -re 's/\.t[0123456789]+//' %s | sort -k 1 | uniq > %s" % (
output_pfam_supported_transcripts_ids, output_pfam_supported_genes_ids)
os.system(remove_transcript_ids_str)
if args.swissprot_db:
print("Annotating peptides(Swissprot database)...")
BLASTp.threads = args.threads
BLASTp.parallel_blastp(output_pep,
args.swissprot_db,
evalue=0.0000001,
output_format=6,
outfile=output_swissprot_blastp_hits,
split_dir="splited_blastp_fasta",
splited_output_dir="splited_blastp_output_dir")
hits_dict = BLASTp.extract_hits_from_tbl_output(
output_swissprot_blastp_hits, output_swissprot_blastp_hits_names)
supported_ids = IdSet(hits_dict.keys())
supported_ids.write(output_swissprot_supported_transcripts_ids)
remove_transcript_ids_str = "sed -re 's/\.t[0123456789]+//' %s | sort -k 1 | uniq > %s" % (
output_swissprot_supported_transcripts_ids,
output_swissprot_supported_genes_ids)
os.system(remove_transcript_ids_str)
for directory in ("splited_blastp_fasta", "splited_blastp_output_dir"):
shutil.rmtree(directory)
if args.database_type == "nucleotide":
MakeBLASTDb.make_nucleotide_db(args.input,
args.name,
mask_file if args.mask else None,
output_file=args.name)
BLASTn.parallel_blastn(args.input,
args.name,
outfile=args.output,
blast_options=args.other_options,
split_dir="splited_fasta",
splited_output_dir="splited_output_dir",
evalue=args.evalue,
output_format=args.output_format,
threads=args.threads,
combine_output_to_single_file=True)
elif args.database_type == "protein":
MakeBLASTDb.make_protein_db(args.input,
args.name,
mask_file if args.mask else None,
output_file=args.name)
BLASTp.parallel_blastp(args.input,
args.name,
outfile=args.output,
blast_options=args.other_options,
split_dir="splited_fasta",
splited_output_dir="splited_output_dir",
evalue=args.evalue,
output_format=args.output_format,
threads=args.threads,
combine_output_to_single_file=True)