def filter_mvf(args): """Main method""" if args.more_help is True: modulehelp() sys.exit() if args.mvf is None and args.test is None: raise RuntimeError("No input file specified with --mvf") if args.out is None and args.test is None: raise RuntimeError("No output file specified with --out") # Establish Input MVF if args.test is not None: ncol = args.test_nchar or len(args.test.split()[1]) else: mvf = MultiVariantFile(args.mvf, 'read') ncol = mvf.metadata['ncol'] # Create Actionset if args.labels: labels = mvf.get_sample_labels()[:] for i in range(len(args.actions)): action = args.actions[i] arr = action.split(':') if arr[0] in ('columns', 'collapsepriority', 'collapsemerge', 'allelegroup', 'notmultigroup'): for j in range(1, len(arr)): arr[j] = ','.join( [str(labels.index(x)) for x in arr[j].split(',')]) args.actions[i] = ':'.join(arr) actionset = build_actionset(args.actions, ncol) # TESTING MODE if args.test: loc, alleles = args.test.split() linefail = False transformed = False # invar = invariant (single character) # refvar (all different than reference, two chars) # onecov (single coverage, + is second character) # onevar (one variable base, + is third character) # full = full alleles (all chars) if args.verbose: print(alleles) linetype = get_linetype(alleles) sys.stdout.write("MVF Encoding type '{}' detected\n".format(linetype)) for actionname, actiontype, actionfunc, actionarg in actionset: sys.stdout.write("Applying action {} ({}): ".format( actionname, actiontype)) if actiontype == 'filter': if not actionfunc(alleles, linetype): linefail = True sys.stdout.write("Filter Fail\n") break else: sys.stdout.write("Filter Pass\n") elif actiontype == 'transform': transformed = True alleles = actionfunc(alleles, linetype) linetype = get_linetype(alleles) if linetype == 'empty': linefail = True sys.stdout.write("Transform removed all alleles\n") break else: sys.stdout.write("Transform result {}\n".format(alleles)) elif actiontype == 'location': loc = loc.split(':') loc[1] = int(loc[1]) if actionfunc(loc) is False: linefail = True sys.stdout.write("Location Fail\n") break else: sys.stdout.write("Location Pass\n") if linefail is False: if transformed: if linetype == 'full': alleles = encode_mvfstring(alleles) if alleles: test_output = "{}\t{}\n".format(loc, alleles) sys.stdout.write("Final output = {}\n".format(test_output)) else: sys.stdout.write("Transform removed all alleles\n") else: sys.stdout.write("No changes applied\n") sys.stdout.write("Final output = {}\n".format(args.test)) sys.exit() # MAIN MODE # Set up file handler outmvf = MultiVariantFile(args.out, 'write', overwrite=args.overwrite) outmvf.metadata = deepcopy(mvf.metadata) # reprocess header if actions are used that filter columns if any(x == y[0] for x in ('columns', 'collapsepriority', 'collapsemerge') for y in actionset): if args.labels: labels = outmvf.metadata['labels'][:] else: labels = [x for x in outmvf.metadata['samples']] for actionname, actiontype, actionfunc, actionarg in actionset: if actionname == 'columns': labels = [labels[x] for x in actionarg[0]] elif actionname in ('collapsepriority', 'collapsemerge'): labels = [ labels[x] for x in range(len(labels)) if x not in actionarg[0][1:] ] if args.labels: oldindices = mvf.get_sample_indices(labels) else: oldindices = labels[:] newsamples = {} for i, _ in enumerate(labels): newsamples[i] = mvf.metadata['samples'][oldindices[i]] outmvf.metadata['samples'] = newsamples.copy() outmvf.metadata['labels'] = labels[:] outmvf.write_data(outmvf.get_header()) # End header editing linebuffer = [] nbuffer = 0 for chrom, pos, allelesets in mvf.iterentries(decode=False): linefail = False transformed = False # invar = invariant (single character) # refvar (all different than reference, two chars) # onecov (single coverage, + is second character) # onevar (one variable base, + is third character) # full = full alleles (all chars) alleles = allelesets[0] linetype = get_linetype(alleles) if linetype == 'empty': continue if args.verbose is True: sys.stdout.write(" {} {}".format(alleles, linetype)) for actionname, actiontype, actionfunc, actionargs in actionset: if actiontype == 'filter': if not actionfunc(alleles, linetype): linefail = True elif actiontype == 'transform': transformed = True alleles = actionfunc(alleles, linetype) linetype = get_linetype(alleles) if linetype == 'empty': linefail = True elif actiontype == 'location': if actionfunc([chrom, pos]) is False: linefail = True if linefail: break if linefail is False: if transformed: if linetype == 'full': alleles = mvf.encode(alleles) if not alleles: linefail = True nbuffer += 1 linebuffer.append((chrom, pos, (alleles, ))) if args.verbose: sys.stdout.write("{}\n".format(alleles)) if nbuffer == args.line_buffer: outmvf.write_entries(linebuffer) linebuffer = [] nbuffer = 0 elif args.verbose: sys.stdout.write("FAIL\n") if linebuffer: outmvf.write_entries(linebuffer) linebuffer = [] return ''
def filter_mvf(args): """Main method""" args.qprint("Running FilterMVF") if args.more_help is True: modulehelp() sys.exit() if args.mvf is None and args.test is None: raise RuntimeError("No input file specified with --mvf") if args.out is None and args.test is None: raise RuntimeError("No output file specified with --out") # Establish Input MVF if args.test is not None: ncol = args.test_nchar or len(args.test.split()[1]) else: mvf = MultiVariantFile(args.mvf, 'read') ncol = mvf.metadata['ncol'] args.qprint("Input MVF read with {} columns.".format(ncol)) # Create Actionset if args.labels: for i in range(len(args.actions)): action = args.actions[i] arr = action.split(':') if arr[0] in ('collapsepriority', 'collapsemerge'): arr[1] = ','.join([ str(mvf.sample_id_to_index[x]) for x in arr[1].split(',')]) if arr[0] in ('columns', 'allelegroup', 'notmultigroup', 'reqsample'): for j in range(1, len(arr)): arr[j] = ','.join([ str(mvf.sample_id_to_index[x]) for x in arr[j].split(',')]) args.actions[i] = ':'.join(arr) removed_columns = set([]) for i in range(len(args.actions)): action = args.actions[i] arr = action.split(':') if arr[0] in ('collapsepriority', 'collapsemerge'): tmp_arr = arr[1][:] arr[1] = ','.join([ str(int(x) - len([y for y in removed_columns if y < int(x)])) for x in arr[1].split(',')]) removed_columns.update([int(x) for x in tmp_arr.split(',')[1:]]) print(arr) print(removed_columns) if arr[0] in ('columns', 'allelegroup', 'notmultigroup', 'reqsample'): for j in range(1, len(arr)): arr[j] = ','.join([ str(int(x) - len([y for y in removed_columns if y < int(x)])) for x in arr[j].split(',')]) args.actions[i] = ':'.join(arr) actionset = build_actionset(args.actions, ncol) args.qprint("Actions established.") args.qprint(actionset) # TESTING MODE if args.test: loc, alleles = args.test.split() linefail = False transformed = False # invar = invariant (single character) # refvar (all different than reference, two chars) # onecov (single coverage, + is second character) # onevar (one variable base, + is third character) # full = full alleles (all chars) if args.verbose: print(alleles) linetype = get_linetype(alleles) sys.stdout.write("MVF Encoding type '{}' detected\n".format(linetype)) for actionname, actiontype, actionfunc, actionarg in actionset: sys.stdout.write("Applying action {} ({}): ".format( actionname, actiontype)) if actiontype == 'filter': if not actionfunc(alleles, linetype): linefail = True sys.stdout.write("Filter Fail\n") break sys.stdout.write("Filter Pass\n") elif actiontype == 'transform': transformed = True alleles = actionfunc(alleles, linetype) linetype = get_linetype(alleles) if linetype == 'empty': linefail = True sys.stdout.write("Transform removed all alleles\n") break sys.stdout.write("Transform result {}\n".format(alleles)) elif actiontype == 'location': loc = loc.split(':') loc[1] = int(loc[1]) if actionfunc(loc) is False: linefail = True sys.stdout.write("Location Fail\n") break sys.stdout.write("Location Pass\n") if linefail is False: if transformed: if linetype == 'full': alleles = encode_mvfstring(alleles) if alleles: test_output = "{}\t{}\n".format(loc, alleles) sys.stdout.write("Final output = {}\n".format( test_output)) else: sys.stdout.write("Transform removed all alleles\n") else: sys.stdout.write("No changes applied\n") sys.stdout.write("Final output = {}\n".format(args.test)) sys.exit() # MAIN MODE # Set up file handler outmvf = MultiVariantFile(args.out, 'write', overwrite=args.overwrite) outmvf.copy_headers_from(mvf) removed_indices = set([]) # reprocess header if actions are used that filter columns if any(x == y[0] for x in ('columns', 'collapsepriority', 'collapsemerge') for y in actionset): for actionname, actiontype, actionfunc, actionarg in actionset: if actionname == 'columns': if args.labels: oldindices = [outmvf.sample_id_to_index[int(x)] for x in actionarg[0]] else: oldindices = [int(x) for x in actionarg[0]] elif actionname in ('collapsepriority', 'collapsemerge'): actionarg[0] = [x - len([y for y in removed_indices if y < x]) for x in actionarg[0]] oldindices = [x for x in outmvf.sample_indices if x not in actionarg[0][1:]] outmvf.sample_ids = outmvf.get_sample_ids(oldindices) outmvf.sample_data = dict( (i, outmvf.sample_data[oldindices[i]]) for i, _ in enumerate(oldindices)) if actionname in ('collapsepriority', 'collapsemerge'): if len(actionarg) == 2: outmvf.sample_data[actionarg[0][0]]['id'] = actionarg[1][0] outmvf.sample_ids[actionarg[0][0]] = actionarg[1][0] outmvf.sample_indices = list(range(len(oldindices))) outmvf.metadata['ncol'] = len(outmvf.sample_indices) outmvf.notes.append(args.command_string) outmvf.write_data(outmvf.get_header()) args.qprint("Output MVF established.") # End header editing linebuffer = [] nbuffer = 0 args.qprint("Processing Entries.") write_total = 0 for chrom, pos, allelesets in mvf.iterentries(decode=False): linefail = False transformed = False # invar = invariant (single character) # refvar (all different than reference, two chars) # onecov (single coverage, + is second character) # onevar (one variable base, + is third character) # full = full alleles (all chars) alleles = allelesets[0] linetype = get_linetype(alleles) if linetype == 'empty': continue if args.verbose is True: sys.stdout.write(" {} {} ".format(alleles, linetype)) for actionname, actiontype, actionfunc, _ in actionset: if actiontype == 'filter': linefail = not actionfunc(alleles, linetype) elif actiontype == 'transform': transformed = True alleles = actionfunc(alleles, linetype) linetype = get_linetype(alleles) linefail = linetype == 'empty' elif actiontype == 'location': linefail = not actionfunc([chrom, pos]) if linefail: break if linefail is False: if transformed: if linetype == 'full': alleles = mvf.encode(alleles) if not alleles: linefail = True nbuffer += 1 linebuffer.append((chrom, pos, (alleles,))) if args.verbose: sys.stdout.write("{}\n".format(alleles)) if nbuffer == args.line_buffer: write_total += args.line_buffer args.qprint("{} entries written. Total written: {}.".format( args.line_buffer, write_total)) outmvf.write_entries(linebuffer) linebuffer = [] nbuffer = 0 elif args.verbose: sys.stdout.write("FAIL\n") if linebuffer: outmvf.write_entries(linebuffer) write_total += len(linebuffer) args.qprint("{} entries written. Total written: {}.".format( args.line_buffer, write_total)) linebuffer = [] return ''
def mvf_join(args): """Main method""" concatmvf = MultiVariantFile(args.out, 'write', overwrite=args.overwrite) # Copy the first file's metadata if args.main_header_file: if args.main_header_file not in args.mvf: raise RuntimeError("{} not found in files".format( args.main_header_file)) else: args.main_header_file = args.mvf.index(args.main_header_file) else: args.main_header_file = 0 first_mvf = MultiVariantFile(args.mvf[args.main_header_file], 'read') concatmvf.metadata = first_mvf.metadata.copy() # Open each MVF file, read headers to make unified header transformers = [] for mvfname in args.mvf: # This will create a dictionary of samples{old:new}, contigs{old:new} transformer = MvfTransformer() mvf = MultiVariantFile(mvfname, 'read') for i, label in enumerate(mvf.get_sample_labels()): if label not in concatmvf.get_sample_labels(): concatmvf.metadata['labels'].append(label) concatmvf.metadata['samples'][ concatmvf.metadata['labels'].index(label)] = { 'label': label } if concatmvf.metadata['labels'].index(label) != i: transformer.set_label( i, concatmvf.metadata['labels'].index(label)) for contigid, contigdata in iter(mvf.metadata['contigs'].items()): if contigdata['label'] not in [ concatmvf.metadata['contigs'][x]['label'] for x in concatmvf.metadata['contigs'] ]: newid = (contigid not in concatmvf.metadata['contigs'] and contigid or concatmvf.get_next_contig_id()) concatmvf.metadata['contigs'][newid] = contigdata else: for concatid, concatdata in ( concatmvf.metadata['contigs'].items()): if contigdata['label'] == concatdata['label']: newid = concatid break if newid != contigid: transformer.set_contig(contigid, newid) transformers.append(transformer) # Write output header concatmvf.write_data(concatmvf.get_header()) # Now loop through each file entries = [] nentries = 0 for ifile, mvfname in enumerate(args.mvf): if not args.quiet: sys.stderr.write("Processing {} ...\n".format(mvfname)) transformer = transformers[ifile] mvf = MultiVariantFile(mvfname, 'read') for contigid, pos, allelesets in mvf.iterentries(decode=False, quiet=args.quiet): if transformer.labels: allelesets = [mvf.decode(x) for x in allelesets] for j, alleles in enumerate(allelesets): allelesets[j] = concatmvf.encode(''.join([ x in transformer.labels and alleles[transformer.labels[x]] or alleles[x] for x in range(len(alleles)) ])) if transformer.contigs: contigid = (contigid in transformer['contigs'] and transformer['contigs'][contigid] or contigid) entries.append((contigid, pos, allelesets)) nentries += 1 if nentries == args.line_buffer: concatmvf.write_entries(entries) entries = [] nentries = 0 if entries: concatmvf.write_entries(entries) entries = [] nentries = 0 if not args.quiet: sys.stderr.write("done\n") return ''
def translate_mvf(args): """Main method""" args.qprint("Running TranslateMVF") if args.gff: args.qprint("Reading and Indexing MVF.") else: args.qprint("Reading MVF.") mvf = MultiVariantFile(args.mvf, 'read', contigindex=bool(args.gff)) if mvf.flavor != 'dna': raise RuntimeError("MVF must be flavor=dna to translate") if args.gff: args.qprint("Processing MVF Index File.") mvf.read_index_file() args.qprint("GFF processing start.") gff_genes, gene_order = parse_gff_exome(args) args.qprint("GFF processed.") outmvf = MultiVariantFile(args.out, 'write', overwrite=args.overwrite) outmvf.copy_headers_from(mvf) outmvf.contig_data = dict( ( i, dict((y, z) for (y, z) in gff_genes[x].items() if y not in ('cds', ))) for (i, x) in enumerate(gene_order)) outmvf.contig_indices = list(range(len(gene_order))) outmvf.contig_ids = [gff_genes[x]['id'] for x in gene_order] outmvf.contig_labels = [gff_genes[x]['label'] for x in gene_order] outmvf.flavor = args.output_data outmvf.metadata.notes.append(args.command_string) outmvf.write_data(outmvf.get_header()) args.qprint("Output MVF Established.") entrybuffer = [] nentry = 0 pos = None if not args.gff: args.qprint("No GFF used, translating sequences as pre-aligned in " "coding frame.") inputbuffer = [] current_contig = '' for contigid, pos, allelesets in mvf.iterentries(decode=False): if current_contig == '': current_contig = contigid[:] if contigid == current_contig: inputbuffer.append((pos, allelesets)) else: for _, amino_acids, alleles in iter_codons( inputbuffer, mvf): if all([x in '-X' for x in amino_acids]): continue if args.output_data == 'protein': entrybuffer.append( (current_contig, pos, (amino_acids,))) else: entrybuffer.append(( current_contig, pos, ( amino_acids, alleles[0], alleles[1], alleles[2]))) nentry += 1 if nentry == args.line_buffer: outmvf.write_entries(entrybuffer) entrybuffer = [] nentry = 0 inputbuffer = [(pos, allelesets)] current_contig = contigid[:] if inputbuffer: for _, amino_acids, alleles in iter_codons( inputbuffer, outmvf): if all([x in '-X' for x in amino_acids]): continue if args.output_data == 'protein': entrybuffer.append( (current_contig, pos, (amino_acids,))) else: entrybuffer.append(( current_contig, pos, ( amino_acids, alleles[0], alleles[1], alleles[2]))) nentry += 1 if nentry == args.line_buffer: outmvf.write_entries(entrybuffer) entrybuffer = [] nentry = 0 else: running_gene_index = -1 for igene, gene in enumerate(gene_order): xcontiglabel = gff_genes[gene]['contig'] xcontig = mvf.get_contig_indices( labels=gff_genes[gene]['contig']) if xcontig is None: print("Warning: contig {} not found".format( gff_genes[gene]['contig'])) xcontigid = mvf.get_contig_ids(indices=xcontig)[0] min_gene_coord = gff_genes[gene]['cds'][0][0] max_gene_coord = gff_genes[gene]['cds'][-1][1] mvf_entries = {} if not igene % 100: args.qprint("Processing gene {} on {}".format( gene, xcontiglabel)) for contigid, pos, allelesets in mvf.itercontigentries( xcontig, decode=False): if pos < min_gene_coord: continue if pos > max_gene_coord: break mvf_entries[pos] = allelesets[0] reverse_strand = gff_genes[gene]['strand'] == '-' coords = [] running_gene_index += 1 for elem in gff_genes[gene]['cds']: coords.extend(list(range(elem[0], elem[1] + 1))) if reverse_strand: coords = coords[::-1] for codoncoord in range(0, len(coords), 3): alleles = tuple(mvf_entries.get(x, '-') for x in coords[codoncoord:codoncoord + 3]) if len(alleles) < 3: alleles = tuple(list(alleles) + ['-'] * (3 - len(alleles))) if all(len(x) == 1 for x in alleles): if reverse_strand: alleles = tuple( MLIB.complement_bases[x] for x in alleles) decoded_alleles = alleles amino_acids = translate_single_codon(''.join(alleles)) else: if reverse_strand is True: decoded_alleles = tuple(tuple(MLIB.complement_bases[y] for y in mvf.decode(x)) for x in alleles) alleles = tuple(outmvf.encode(''.join(x)) for x in decoded_alleles) else: decoded_alleles = tuple(mvf.decode(x) for x in alleles) amino_acids = tuple(translate_single_codon(''.join(x)) for x in zip(*decoded_alleles)) amino_acids = outmvf.encode(''.join(amino_acids)) if args.output_data == 'protein': entrybuffer.append(( ( xcontigid if args.retain_contigs else running_gene_index ), ( coords[codoncoord] if args.retain_coords else codoncoord ), ( amino_acids, ) )) elif args.output_data == 'codon': entrybuffer.append(( ( xcontigid if args.retain_contigs else running_gene_index ), ( coords[codoncoord] if args.retain_coords else codoncoord ), ( amino_acids, alleles[0], alleles[1], alleles[2] ) )) elif args.output_data == 'dna': for j, elem in enumerate( range(codoncoord, min(codoncoord + 3, len(coords)))): entrybuffer.append(( ( xcontigid if args.retain_contigs else running_gene_index ), ( coords[elem] if args.retain_coords else elem + 1 ), ( alleles[j], ) )) nentry += 1 if nentry >= args.line_buffer: args.qprint("Writing a block of {} entries.".format( args.line_buffer)) outmvf.write_entries(entrybuffer) entrybuffer = [] nentry = 0 if entrybuffer: outmvf.write_entries(entrybuffer) entrybuffer = [] nentry = 0 return ''
def legacy_translate_mvf(args): """Main method""" args.qprint("Running LegacyTranslateMVF") if args.gff: args.qprint("Reading and Indexing MVF.") else: args.qprint("Reading MVF.") mvf = MultiVariantFile(args.mvf, 'read', contigindex=bool(args.gff)) if mvf.flavor != 'dna': raise RuntimeError("MVF must be flavor=dna to translate") if args.gff: args.qprint("Processing MVF Index File.") mvf.read_index_file() args.qprint("GFF processing start.") gff = parse_gff_legacy_translate( args.gff, args, parent_gene_pattern=args.parent_gene_pattern) args.qprint("GFF processed.") outmvf = MultiVariantFile(args.out, 'write', overwrite=args.overwrite) outmvf.copy_headers_from(mvf) outmvf.flavor = args.output_data outmvf.write_data(outmvf.get_header()) args.qprint("Output MVF Established.") entrybuffer = [] nentry = 0 pos = None if not args.gff: args.qprint("No GFF used, translating sequences as pre-aligned in " "coding frame.") inputbuffer = [] current_contig = '' for contigid, pos, allelesets in mvf.iterentries(decode=False): if current_contig == '': current_contig = contigid[:] if contigid == current_contig: inputbuffer.append((pos, allelesets)) else: for _, amino_acids, alleles in iter_codons( inputbuffer, mvf): if all([x in '-X' for x in amino_acids]): continue if args.output_data == 'protein': entrybuffer.append( (current_contig, pos, (amino_acids,))) else: entrybuffer.append(( current_contig, pos, ( amino_acids, alleles[0], alleles[1], alleles[2]))) nentry += 1 if nentry == args.line_buffer: outmvf.write_entries(entrybuffer) entrybuffer = [] nentry = 0 inputbuffer = [(pos, allelesets)] current_contig = contigid[:] if inputbuffer: for _, amino_acids, alleles in iter_codons( inputbuffer, outmvf): if all([x in '-X' for x in amino_acids]): continue if args.output_data == 'protein': entrybuffer.append( (current_contig, pos, (amino_acids,))) else: entrybuffer.append(( current_contig, pos, ( amino_acids, alleles[0], alleles[1], alleles[2]))) nentry += 1 if nentry == args.line_buffer: outmvf.write_entries(entrybuffer) entrybuffer = [] nentry = 0 else: args.qprint("Indexing GFF gene names.") # mvfid_to_gffname = outmvf.get_contig_reverse_dict() for xcontig in outmvf.get_contig_indices(): mvf_entries = {} xcontiglabel = outmvf.get_contig_labels(indices=xcontig)[0] xcontigid = outmvf.get_contig_ids(indices=xcontig)[0] if xcontiglabel not in gff: if args.verbose: print( ("No entries in GFF, " "skipping contig: index:{} id:{} label:{}").format( xcontig, xcontigid, xcontiglabel)) continue if not xcontig % 100: args.qprint("Processing contig: {} {}".format( xcontigid, xcontiglabel)) for contigid, pos, allelesets in mvf.itercontigentries( xcontig, decode=False): mvf_entries[pos] = allelesets[0] for coords in sorted(gff[xcontiglabel]): reverse_strand = coords[3] == '-' alleles = (tuple(mvf_entries.get(x, '-') for x in coords[2::-1]) if reverse_strand is True else tuple(mvf_entries.get(x, '-') for x in coords[0:3])) if all(len(x) == 1 for x in alleles): if reverse_strand: alleles = tuple( MLIB.complement_bases[x] for x in alleles) decoded_alleles = alleles amino_acids = translate_single_codon(''.join(alleles)) else: if reverse_strand is True: decoded_alleles = tuple(tuple(MLIB.complement_bases[y] for y in mvf.decode(x)) for x in alleles) alleles = tuple(outmvf.encode(''.join(x)) for x in decoded_alleles) else: decoded_alleles = tuple(mvf.decode(x) for x in alleles) amino_acids = tuple(translate_single_codon(''.join(x)) for x in zip(*decoded_alleles)) # print("aminx", amino_acids) amino_acids = outmvf.encode(''.join(amino_acids)) # if all(x in '-X' for x in amino_acids): # continue # print("amino", amino_acids) # print("translated", amino_acids, alleles) if args.output_data == 'protein': entrybuffer.append((xcontig, coords[0], (amino_acids,))) else: entrybuffer.append(( xcontigid, coords[0], ( amino_acids, alleles[0], alleles[1], alleles[2]))) nentry += 1 if nentry >= args.line_buffer: args.qprint("Writing a block of {} entries.".format( args.line_buffer)) outmvf.write_entries(entrybuffer) entrybuffer = [] nentry = 0 if entrybuffer: outmvf.write_entries(entrybuffer) entrybuffer = [] nentry = 0 return ''
def translate_mvf(args): """Main method""" mvf = MultiVariantFile(args.mvf, 'read') if mvf.flavor != 'dna': raise RuntimeError("MVF must be flavor=dna to translate") if args.gff: gff = parse_gff_translate(args.gff, args) if not args.quiet: print("gff_processed") outmvf = MultiVariantFile(args.out, 'write', overwrite=args.overwrite) outmvf.metadata = deepcopy(mvf.metadata) outmvf.flavor = args.output_data outmvf.write_data(outmvf.get_header()) entrybuffer = [] nentry = 0 if not args.gff: inputbuffer = [] current_contig = '' for contigid, pos, allelesets in mvf.iterentries(decode=False): if current_contig == '': current_contig = contigid[:] if contigid == current_contig: inputbuffer.append((pos, allelesets)) else: for _, amino_acids, alleles in iter_codons(inputbuffer, mvf): if all([x in '-X' for x in amino_acids]): continue if args.output_data == 'protein': entrybuffer.append( (current_contig, pos, (amino_acids, ))) else: entrybuffer.append( (current_contig, pos, (amino_acids, alleles[0], alleles[1], alleles[2]))) nentry += 1 if nentry == args.line_buffer: outmvf.write_entries(entrybuffer) entrybuffer = [] nentry = 0 inputbuffer = [(pos, allelesets)] current_contig = contigid[:] if inputbuffer: for _, amino_acids, alleles in iter_codons(inputbuffer, mvf): if all([x in '-X' for x in amino_acids]): continue if args.output_data == 'protein': entrybuffer.append((current_contig, pos, (amino_acids, ))) else: entrybuffer.append( (current_contig, pos, (amino_acids, alleles[0], alleles[1], alleles[2]))) nentry += 1 if nentry == args.line_buffer: outmvf.write_entries(entrybuffer) entrybuffer = [] nentry = 0 else: mvf_entries = {} for contigid, pos, allelesets in mvf.iterentries(decode=False): if contigid not in mvf_entries: mvf_entries[contigid] = {} mvf_entries[contigid][pos] = allelesets[0] for contigname in sorted(gff): contigid = mvf.get_contig_ids(labels=contigname)[0] for coords in sorted(gff[contigname]): reverse_strand = False if coords[3] == '-': reverse_strand = True alleles = [ mvf_entries[contigid].get(x, '-') for x in coords[2::-1] ] else: alleles = [ mvf_entries[contigid].get(x, '-') for x in coords[0:3] ] if all(len(x) == 1 for x in alleles): if reverse_strand: alleles = [MLIB.complement_bases[x] for x in alleles] decoded_alleles = alleles amino_acids = translate(''.join(alleles))[0] else: if reverse_strand: decoded_alleles = [[ MLIB.complement_bases[y] for y in mvf.decode(x) ] for x in alleles] alleles = [ mvf.encode(''.join(x)) for x in decoded_alleles ] else: decoded_alleles = [mvf.decode(x) for x in alleles] amino_acids = [ translate(''.join(x)) for x in zip(*decoded_alleles) ] amino_acids = mvf.encode(''.join( [x[0] for x in amino_acids])) if all([x in '-X' for x in amino_acids]): continue if args.output_data == 'protein': entrybuffer.append((contigid, coords[0], (amino_acids, ))) else: entrybuffer.append( (contigid, coords[0], (amino_acids, alleles[0], alleles[1], alleles[2]))) nentry += 1 if nentry == args.line_buffer: outmvf.write_entries(entrybuffer) entrybuffer = [] nentry = 0 if entrybuffer: outmvf.write_entries(entrybuffer) entrybuffer = [] nentry = 0 return ''