Mercurial > repos > rakesh4osdd > clsi_profile

comparison asist_dynamic.py @ 9:b3c01b790314 draft

Find changesets by keywords (author, files, the commit message), revision number or hash, or revset expression.
"planemo upload for repository https://github.com/rakesh4osdd/asist/tree/master commit f590c3b1d71a9b8f2030909fa488b4ac0c3caed8"
author rakesh4osdd
date Wed, 30 Jun 2021 06:37:12 +0000
parents
children
comparison
equal deleted inserted replaced
8:c89ee0059c70 9:b3c01b790314
1 #!/usr/bin/env python
2 # coding: utf-8
3
4 # In[1309]:
5
6
7 #ASIST program for phenotype based on Antibiotics profile
8 # create a profile based on selected antibiotics only
9 # rakesh4osdd@gmail.com, 14-June-2021
10
11
12 # In[1]:
13
14
15 import pandas as pd
16 import sys
17 import os
18 from collections import Counter
19
20
21 # In[176]:
22
23
24 input_file=sys.argv[1]
25 output_file=sys.argv[2]
26 #input_file='test-data/asist_input.csv'
27 #output_file='test-data/asist_output.csv'
28
29
30 # In[177]:
31
32
33 # strain_profile to phenotype condition
34 def s_phen(sus,res,intm,na,pb_sus):
35 if (sus>0 and res==0 and na>=0):
36 #print('Possible Susceptible')
37 phen='Possible Susceptible'
38 elif (sus>=0 and 3<=res<7 and na>=0 and pb_sus==0):
39 #print('Possible MDR')
40 phen='Possible MDR'
41 elif (sus>=0 and 7<=res<9 and na>=0 and pb_sus==0):
42 #print('Possible XDR')
43 phen='Possible XDR'
44 #special cases
45 elif (sus>=1 and res>0 and na>=0 and pb_sus==1):
46 #print('Possible XDR')
47 phen='Possible XDR'
48 #special cases
49 elif (sus>0 and res==9 and na>=0):
50 #print('Possible XDR')
51 phen='Possible XDR'
52 elif (sus==0 and res==9 and na>=0):
53 #print('Possible TDR')
54 phen='Possible TDR'
55 else:
56 #print('Strain could not be classified')
57 phen='Strain could not be classified ('+ str(intm)+' | ' + str(na) +')'
58 return(phen)
59
60 #print(s_phen(1,9,0,0))
61
62
63 # In[178]:
64
65
66 # define Antibiotic groups as per antibiotic of CLSI breakpoints MIC
67 #Aminoglycoside
68 cat1=['Amikacin','Tobramycin','Gentamycin','Netilmicin']
69 #Beta-lactams- Carbapenems
70 cat2=['Imipenem','Meropenam','Doripenem']
71 #Fluoroquinolone
72 cat3=['Ciprofloxacin','Levofloxacin']
73 #Beta-lactam inhibitor
74 cat4=['Piperacillin/tazobactam','Ticarcillin/clavulanicacid']
75 #Cephalosporin
76 cat5=['Cefotaxime','Ceftriaxone','Ceftazidime','Cefepime']
77 #Sulfonamides
78 cat6=['Trimethoprim/sulfamethoxazole']
79 #Penicillins/beta-lactamase
80 cat7=['Ampicillin/sulbactam']
81 #Polymyxins
82 cat8=['Colistin','Polymyxinb']
83 #Tetracycline
84 cat9=['Tetracycline','Doxicycline','Minocycline']
85
86 def s_profiler(pd_series):
87 #print(type(pd_series),'\n', pd_series)
88 #create a dictionary of dataframe series
89 cats={'s1':cat1,'s2':cat2,'s3':cat3,'s4':cat4,'s5':cat5,'s6':cat6,'s7':cat7,'s8':cat8,'s9':cat9}
90 # find the antibiotics name in input series
91 for cat in cats:
92 #print(cats[cat])
93 cats[cat]=pd_series.filter(cats[cat])
94 #print(cats[cat])
95 #define res,sus,intm,na,pb_sus
96 res=0
97 sus=0
98 intm=0
99 na=0
100 pb_sus=0
101 # special case of 'Polymyxin b' for its value
102 if 'Polymyxinb' in pd_series:
103 ctp=cats['s8']['Polymyxinb'].strip().lower()
104 if ctp == 'susceptible':
105 pb_sus=1
106 #print((ctp,p_sus))
107 # check all categories
108 for cat in cats:
109 #ctp=cats['s8'].iloc[i:i+1].stack().value_counts().to_dict()
110 #print(ctp)
111 # Pandas series
112 ct=cats[cat].value_counts().to_dict()
113 #print(ct)
114 # remove whitespace and convert to lowercase words
115 ct = {k.strip().lower(): v for k, v in ct.items()}
116 #print(ct)
117 k=Counter(ct)
118 #j=Counter(ct)+Counter(j)
119 #print(j)
120 # category wise marking
121 if k['resistant']>=1:
122 res=res+1
123 if k['susceptible']>=1:
124 sus=sus+1
125 if k['intermediate']>=1:
126 intm=intm+1
127 if k['na']>=1:
128 na=na+1
129 #print(sus,res,intm,na,pb_sus)
130 #print(s_phen(sus,res,intm,na,pb_sus))
131 return(s_phen(sus,res,intm,na,pb_sus))
132
133
134 # In[179]:
135
136
137 #input_file='input2.csv_table.csv'
138 #output_file=input_file+'_output.txt'
139 strain_profile=pd.read_csv(input_file, sep=',',na_filter=False,skipinitialspace = True)
140
141
142 # In[180]:
143
144
145 old_strain_name=strain_profile.columns[0]
146 new_strain_name=old_strain_name.capitalize().strip().replace(' ', '')
147
148
149 # In[181]:
150
151
152 # make header capitalization, remove leading,lagging, and multiple whitespace for comparision
153 strain_profile.columns=strain_profile.columns.str.capitalize().str.strip().str.replace('\s+', '', regex=True)
154 #print(strain_profile.columns)
155 #strain_profile.head()
156 #strain_profile.columns
157
158
159 # In[182]:
160
161
162 # add new column in dataframe on second position
163 strain_profile.insert(1, 'Strain phenotype','')
164 #strain_profile.head()
165
166
167 # In[183]:
168
169
170 strain_profile['Strain phenotype'] = strain_profile.apply(lambda x: (s_profiler(x)), axis=1)
171
172
173 # In[184]:
174
175
176 #strain_profile.head()
177
178
179 # In[185]:
180
181
182 #rename headers for old name
183 strain_profile=strain_profile.rename(columns = {new_strain_name:old_strain_name, 'Ticarcillin/clavulanicacid':'Ticarcillin/ clavulanic acid','Piperacillin/tazobactam':'Piperacillin/ tazobactam','Trimethoprim/sulfamethoxazole': 'Trimethoprim/ sulfamethoxazole','Ampicillin/sulbactam':'Ampicillin/ sulbactam', 'Polymyxinb': 'Polymyxin B'} )
184
185
186 # In[186]:
187
188
189 #strain_profile.columns
190
191
192 # In[187]:
193
194
195 #strain_profile
196
197
198 # In[188]:
199
200
201 strain_profile.to_csv(output_file,na_rep='NA',index=False)
202
203
204 # In[189]:
205
206
207 # Open a file with access mode 'a'
208 with open(output_file, "a") as file_object:
209 # Append 'hello' at the end of file
210 file_object.write("Note: \n1. 'MDR': Multidrug-resistant, 'XDR': Extensively drug-resistant, 'TDR':totally drug resistant, NA': Data Not Available.\n2. 'Strain could not be classified' numbers follow the format as ('Number of antibiotics categories count as Intermediate' | 'Number of antibiotics categories count as NA')")
211
212
213 # In[ ]:
214
215
216
217