saving from work

mcsm_analysis/pyrazinamide/scripts/generate_mut_sequences.py

@@ -9,6 +9,7 @@ Created on Tue Jun 25 08:46:36 2019
 # load libraries
 import os
 import pandas as pd
+import numpy as np
 from Bio import SeqIO
 ############################################
 #********************************************************************
@@ -47,7 +48,7 @@ print("Input file is:", infile_meta_data)
 #=======
 # output 
 #=======
-outdir =   'git/Data/pyrazinamide/output'
+outdir = 'git/Data/pyrazinamide/output'
 # filenames in respective sections
 
 ################## end of variable assignment for input and output files
@@ -86,7 +87,7 @@ print("new length:", len(my_fasta))
 #############
 # read mutant_info file and extract cols with positions and mutant_info
 # This should be all samples with pncA muts
-#my_data = pd.read_csv('mcsm_complex1_normalised.csv') #335, 15
+#my_data = pd.read_csv('mcsm_complex1_normalised.csv') 
 my_data = pd.read_csv(infile_meta_data)
 list(my_data.columns)
 #my_data['OR'].value_counts()
@@ -95,12 +96,34 @@ list(my_data.columns)
 #FIXME: You need a better way to identify this
 # ideally this file should not contain any non_struc pos
 # remove positions not in the structure
-my_data = my_data[my_data.position != ns_pos_o] #3092, 22
+my_data = my_data[my_data.position != ns_pos_o] 
 
 # if multiple positions, then try the example below;
 # https://stackoverflow.com/questions/29017525/deleting-rows-based-on-multiple-conditions-python-pandas
 #df = df[(df.one > 0) | (df.two > 0) | (df.three > 0) & (df.four < 1)]
 
+# count mutations per sample
+mut_info = my_data[['id', 'Mutationinformation', 'wild_type', 'position', 'mutant_type']]
+
+# test
+foo = mut_info[mut_info.Mutationinformation.str.contains('C72Y')]
+ 
+foo = mut_info.pivot_table(values = ['Mutationinformation']
+                           , index = ['Mutationinformation', 'id']
+#                           , columns = 
+                           , aggfunc = 'count')
+
+# table
+foo_tab = mut_info.pivot_table(values = ['Mutationinformation']
+#                           , index = ['Mutationinformation']
+                           , columns = ['id', 'Mutationinformation']
+                           , aggfunc = 'count'
+#                           , margins = True)
+                           )
+foo_tab.stack('id')
+
+mut_info.to_csv('mutinfo.csv')
+
 mut_info1 = my_data[['position', 'mutant_type']]
 #%%
 ################