forked from jenniferlu717/KrakenTools
-
Notifications
You must be signed in to change notification settings - Fork 0
/
combine_mpa.py
executable file
·142 lines (133 loc) · 5.43 KB
/
combine_mpa.py
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
61
62
63
64
65
66
67
68
69
70
71
72
73
74
75
76
77
78
79
80
81
82
83
84
85
86
87
88
89
90
91
92
93
94
95
96
97
98
99
100
101
102
103
104
105
106
107
108
109
110
111
112
113
114
115
116
117
118
119
120
121
122
123
124
125
126
127
128
129
130
131
132
133
134
135
136
137
138
139
140
141
142
#!/usr/bin/env python
####################################################################
#combine_mpa.py converts multiple outputs from kreport2mpa.py
#Copyright (C) 2020 Jennifer Lu, jennifer.lu717@gmail.com
#This file is part of KrakenTools.
#KrakenTools is free software; you can redistribute it and/or modify
#it under the terms of the GNU General Public License as published by
#the Free Software Foundation; either version 3 of the license, or
#(at your option) any later version.
#This program is distributed in the hope that it will be useful,
#but WITHOUT ANY WARRANTY; without even the implied warranty of
#MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the
#GNU General Public License for more details.
#You should have received a copy of the GNU General Public License
#along with this program; if not, see <http://www.gnu.org/licenses/>.
####################################################################
#Jennifer Lu, jlu26@jhmi.edu
#Updated: 07/12/2020
#
#This program reads multiple files in the
#an mpa-format (MetaPhlAn) style report (as output from kreport2mpa.py).
#Each line represents a possible taxon classification. The first column is lists the
#domain, kingdom, phyla, etc, leading up to each taxon.
#The levels are separated by the | delimiter, with the type of
#level specified before each name with a single letter and underscore
#(d_ for domain, k_ for kingdom, etc).
#The second column is the number of reads classified within
#that taxon's subtree.
#
#Input files:
# - MetaPhlAn format (mpa-format) files with two columns
# - All files must be generated from the same database, with the same
# options from kreport2krona.py or errors may occur
#
#Input Parameters to Specify [OPTIONAL]:
# - header_line = prints a header line in mpa-report
# [Default: no header]
# - intermediate-ranks = includes non-traditional taxon levels
# (traditional levels: domain, kingdom, phylum, class, order,
# family, genus, species)
# [Default: no intermediate ranks]
#Output file format (tab-delimited)
# - Taxonomy tree levels |-delimited, with level type [d,k,p,c,o,f,g,s,x]
# - Number of reads within subtree of the specified level
#
#Methods
# - main
#
import os, sys, argparse
#Main method
def main():
#Parse arguments
parser = argparse.ArgumentParser()
parser.add_argument('-i', '--input', required=True,
nargs='+', dest='in_files',
help='Input files for this program (files generated by kreport2mpa.py)')
parser.add_argument('-o', '--output', required=True,
dest='o_file', help='Single mpa-report file name')
args=parser.parse_args()
#Process each file
samples = {} #Map number to name
sample_count = 0
values = {} #Map taxon tree to sample to number
parent2child = {}
toparse = []
sys.stdout.write(" Number of files to parse: %i\n" % len(args.in_files))
for in_file in args.in_files:
i_file = open(in_file,'r')
sample_count += 1
sample_name = "Sample #" + str(sample_count)
for line in i_file:
#Check for header line
if line[0] == "#":
sample_name = line.strip().split('\t')[-1]
continue
#Otherwise
[classification, val] = line.strip().split('\t')
#Check for parents
split_vals = classification.split("|")
curr_parent = ''
for i in range(0,len(split_vals)):
test_val = "|".join(split_vals[0:i])
if test_val in values:
curr_parent = test_val
#No parent
if curr_parent == '':
if classification not in values:
toparse.append(classification)
#Most specific parent found
if curr_parent != '':
if curr_parent not in parent2child:
parent2child[curr_parent] = []
if classification not in parent2child[curr_parent]:
parent2child[curr_parent].append(classification)
#Save classification to value map
if classification not in values:
values[classification] = {}
values[classification][sample_count] = val
#Save sample name
samples[sample_count] = sample_name
sys.stdout.write(" Number of classifications to write: %i\n" % len(values))
sys.stdout.write("\t%i classifications printed" % 0)
#Write header
o_file = open(args.o_file, 'w')
o_file.write("#Classification")
for i in range(1, sample_count+1):
o_file.write("\t" + samples[i])
o_file.write("\n")
#Write each line
parsed = {}
count_c = 0
while len(toparse) > 0:
curr_c = toparse.pop(0)
#Add all children to stack
if curr_c in parent2child:
for child in parent2child[curr_c]:
toparse.insert(0, child)
#For the current classification, print per sample
o_file.write(curr_c)
for i in range(1,sample_count + 1):
if i in values[curr_c]:
o_file.write("\t" + values[curr_c][i])
else:
o_file.write("\t0")
o_file.write("\n")
count_c += 1
sys.stdout.write("\r\t%i classifications printed" % count_c)
sys.stdout.flush()
o_file.close()
sys.stdout.write("\r\t%i classifications printed\n" % count_c)
sys.stdout.flush()
if __name__ == "__main__":
main()