Mercurial > repos > bebatut > combine_metaphlan2_humann2
comparison combine_metaphlan_humann.py @ 3:01ac9954c27f draft
planemo upload for repository https://github.com/bgruening/galaxytools/tree/master/tools/combine_metaphlan2_humann2 commit b84cbcbffa25c55acd1a31df76e4a4f78772cbd7
| author | bgruening |
|---|---|
| date | Thu, 20 Jul 2023 10:07:12 +0000 |
| parents | |
| children |
comparison
equal
deleted
inserted
replaced
| 2:fdfb35745104 | 3:01ac9954c27f |
|---|---|
| 1 #!/usr/bin/env python | |
| 2 # -*- coding: utf-8 -*- | |
| 3 | |
| 4 import argparse | |
| 5 | |
| 6 | |
| 7 def extract_clade_abundance(metaphlan_fp): | |
| 8 clade_abund = {} | |
| 9 with open(metaphlan_fp, "r") as metaphlan_f: | |
| 10 is_metaphlan_v4 = False | |
| 11 for line in metaphlan_f.readlines(): | |
| 12 if 'SGB' in line: | |
| 13 # New versions of metaphlan against a recent DB contain a header line with DB name, which contains SGB | |
| 14 is_metaphlan_v4 = True | |
| 15 if line.find("g__") == -1: | |
| 16 continue | |
| 17 | |
| 18 split_line = line[:-1].split("\t") | |
| 19 taxo = split_line[0] | |
| 20 if is_metaphlan_v4: | |
| 21 # Column order in new metaphlan versions: | |
| 22 # clade_name NCBI_tax_id relative_abundance additional_species | |
| 23 abundance = split_line[2] | |
| 24 else: | |
| 25 # Column order in the old metaphlan versions: | |
| 26 # clade_name relative_abundance coverage average_genome_length_in_the_clade estimated_number_of_reads_from_the_clade | |
| 27 abundance = split_line[1] | |
| 28 | |
| 29 genus = taxo[(taxo.find("g__") + 3):] | |
| 30 if genus.find("|") != -1: | |
| 31 genus = genus[: (genus.find("|"))] | |
| 32 clade_abund.setdefault(genus, {"abundance": 0, "species": {}}) | |
| 33 if taxo.find("t__") != -1: | |
| 34 continue | |
| 35 elif taxo.find("s__") != -1: | |
| 36 species = taxo[(taxo.find("s__") + 3):] | |
| 37 clade_abund[genus]["species"].setdefault(species, abundance) | |
| 38 else: | |
| 39 clade_abund[genus]["abundance"] = abundance | |
| 40 return clade_abund | |
| 41 | |
| 42 | |
| 43 def compute_overall_abundance(humann_fp): | |
| 44 overall_abundance = 0 | |
| 45 with open(humann_fp, "r") as humann_f: | |
| 46 for line in humann_f.readlines(): | |
| 47 if line.find("|") != -1 or line.startswith("#"): | |
| 48 continue | |
| 49 split_line = line[:-1].split("\t") | |
| 50 overall_abundance += float(split_line[1]) | |
| 51 return overall_abundance | |
| 52 | |
| 53 | |
| 54 def format_characteristic_name(name): | |
| 55 formatted_n = name | |
| 56 formatted_n = formatted_n.replace("/", " ") | |
| 57 formatted_n = formatted_n.replace("-", " ") | |
| 58 formatted_n = formatted_n.replace("'", "") | |
| 59 if formatted_n.find("(") != -1 and formatted_n.find(")") != -1: | |
| 60 open_bracket = formatted_n.find("(") | |
| 61 close_bracket = formatted_n.find(")") + 1 | |
| 62 formatted_n = formatted_n[:open_bracket] + formatted_n[close_bracket:] | |
| 63 return formatted_n | |
| 64 | |
| 65 | |
| 66 def combine_metaphlan_humann(args): | |
| 67 clade_abund = extract_clade_abundance(args.metaphlan_fp) | |
| 68 overall_abund = compute_overall_abundance(args.humann_fp) | |
| 69 | |
| 70 with open(args.output_fp, "w") as output_f: | |
| 71 s = "genus\tgenus_abundance\tspecies\tspecies_abundance\t" | |
| 72 s = "%s\t%s_id\t%s_name\t%s_abundance\n" % (s, args.type, args.type, args.type) | |
| 73 output_f.write(s) | |
| 74 with open(args.humann_fp, "r") as humann_f: | |
| 75 for line in humann_f.readlines(): | |
| 76 if line.find("|") == -1: | |
| 77 continue | |
| 78 | |
| 79 split_line = line[:-1].split("\t") | |
| 80 abundance = 100 * float(split_line[1]) / overall_abund | |
| 81 annotation = split_line[0].split("|") | |
| 82 charact = annotation[0].split(":") | |
| 83 charact_id = charact[0] | |
| 84 char_name = "" | |
| 85 if len(charact) > 1: | |
| 86 char_name = format_characteristic_name(charact[-1]) | |
| 87 taxo = annotation[1].split(".") | |
| 88 | |
| 89 if taxo[0] == "unclassified": | |
| 90 continue | |
| 91 genus = taxo[0][3:] | |
| 92 species = taxo[1][3:] | |
| 93 | |
| 94 if genus not in clade_abund: | |
| 95 print("no %s found in %s" % (genus, args.metaphlan_fp)) | |
| 96 continue | |
| 97 if species not in clade_abund[genus]["species"]: | |
| 98 print( | |
| 99 "No %s found in %s for % s" | |
| 100 % (species, args.metaphlan_fp, genus) | |
| 101 ) | |
| 102 continue | |
| 103 | |
| 104 s = "%s\t%s\t" % (genus, clade_abund[genus]["abundance"]) | |
| 105 s += "%s\t%s\t" % (species, clade_abund[genus]["species"][species]) | |
| 106 s += "%s\t%s\t%s\n" % (charact_id, char_name, abundance) | |
| 107 output_f.write(s) | |
| 108 | |
| 109 | |
| 110 if __name__ == "__main__": | |
| 111 parser = argparse.ArgumentParser() | |
| 112 parser.add_argument("--humann_fp", required=True) | |
| 113 parser.add_argument("--metaphlan_fp", required=True) | |
| 114 parser.add_argument("--output_fp", required=True) | |
| 115 parser.add_argument("--type", required=True, choices=["gene_families", "pathways"]) | |
| 116 args = parser.parse_args() | |
| 117 | |
| 118 combine_metaphlan_humann(args) |
