774 lines
31 KiB
Python
774 lines
31 KiB
Python
#!/usr/bin/env python3
|
|
# -*- coding: utf-8 -*-
|
|
"""
|
|
Created on Sun Mar 6 13:41:54 2022
|
|
|
|
@author: tanu
|
|
"""
|
|
def setvars(gene,drug):
|
|
#https://stackoverflow.com/questions/51695322/compare-multiple-algorithms-with-sklearn-pipeline
|
|
import os, sys
|
|
import pandas as pd
|
|
import numpy as np
|
|
print(np.__version__)
|
|
print(pd.__version__)
|
|
import pprint as pp
|
|
from copy import deepcopy
|
|
from collections import Counter
|
|
from sklearn.impute import KNNImputer as KNN
|
|
from imblearn.over_sampling import RandomOverSampler
|
|
from imblearn.under_sampling import RandomUnderSampler
|
|
from imblearn.over_sampling import SMOTE
|
|
from sklearn.datasets import make_classification
|
|
from imblearn.combine import SMOTEENN
|
|
from imblearn.combine import SMOTETomek
|
|
|
|
from imblearn.over_sampling import SMOTENC
|
|
from imblearn.under_sampling import EditedNearestNeighbours
|
|
from imblearn.under_sampling import RepeatedEditedNearestNeighbours
|
|
|
|
from sklearn.metrics import make_scorer, confusion_matrix, accuracy_score, balanced_accuracy_score, precision_score, average_precision_score, recall_score
|
|
from sklearn.metrics import roc_auc_score, roc_curve, f1_score, matthews_corrcoef, jaccard_score, classification_report
|
|
|
|
from sklearn.model_selection import train_test_split, cross_validate, cross_val_score
|
|
from sklearn.model_selection import StratifiedKFold,RepeatedStratifiedKFold, RepeatedKFold
|
|
|
|
from sklearn.pipeline import Pipeline, make_pipeline
|
|
#%% GLOBALS
|
|
rs = {'random_state': 42}
|
|
njobs = {'n_jobs': 10}
|
|
|
|
scoring_fn = ({ 'mcc' : make_scorer(matthews_corrcoef)
|
|
, 'accuracy' : make_scorer(accuracy_score)
|
|
, 'fscore' : make_scorer(f1_score)
|
|
, 'precision' : make_scorer(precision_score)
|
|
, 'recall' : make_scorer(recall_score)
|
|
, 'roc_auc' : make_scorer(roc_auc_score)
|
|
, 'jcc' : make_scorer(jaccard_score)
|
|
})
|
|
|
|
skf_cv = StratifiedKFold(n_splits = 10
|
|
#, shuffle = False, random_state= None)
|
|
, shuffle = True,**rs)
|
|
|
|
rskf_cv = RepeatedStratifiedKFold(n_splits = 10
|
|
, n_repeats = 3
|
|
, **rs)
|
|
|
|
mcc_score_fn = {'mcc': make_scorer(matthews_corrcoef)}
|
|
jacc_score_fn = {'jcc': make_scorer(jaccard_score)}
|
|
|
|
#%% FOR LATER: Combine ED logo data
|
|
###########################################################################
|
|
rs = {'random_state': 42}
|
|
njobs = {'n_jobs': 10}
|
|
homedir = os.path.expanduser("~")
|
|
|
|
geneL_basic = ['pnca']
|
|
geneL_na = ['gid']
|
|
geneL_na_ppi2 = ['rpob']
|
|
geneL_ppi2 = ['alr', 'embb', 'katg']
|
|
|
|
#num_type = ['int64', 'float64']
|
|
num_type = ['int16', 'int32', 'int64', 'float16', 'float32', 'float64']
|
|
cat_type = ['object', 'bool']
|
|
|
|
#==============
|
|
# directories
|
|
#==============
|
|
datadir = homedir + '/git/Data/'
|
|
indir = datadir + drug + '/input/'
|
|
outdir = datadir + drug + '/output/'
|
|
|
|
#=======
|
|
# input
|
|
#=======
|
|
|
|
#---------
|
|
# File 1
|
|
#---------
|
|
infile_ml1 = outdir + gene.lower() + '_merged_df3.csv'
|
|
#infile_ml2 = outdir + gene.lower() + '_merged_df2.csv'
|
|
|
|
my_features_df = pd.read_csv(infile_ml1, index_col = 0)
|
|
my_features_df = my_features_df .reset_index(drop = True)
|
|
my_features_df.index
|
|
|
|
my_features_df.dtypes
|
|
mycols = my_features_df.columns
|
|
|
|
#---------
|
|
# File 2
|
|
#---------
|
|
infile_aaindex = outdir + 'aa_index/' + gene.lower() + '_aa.csv'
|
|
aaindex_df = pd.read_csv(infile_aaindex, index_col = 0)
|
|
aaindex_df.dtypes
|
|
|
|
#-----------
|
|
# check for non-numerical columns
|
|
#-----------
|
|
if any(aaindex_df.dtypes==object):
|
|
print('\naaindex_df contains non-numerical data')
|
|
|
|
aaindex_df_object = aaindex_df.select_dtypes(include = cat_type)
|
|
print('\nTotal no. of non-numerial columns:', len(aaindex_df_object.columns))
|
|
|
|
expected_aa_ncols = len(aaindex_df.columns) - len(aaindex_df_object.columns)
|
|
|
|
#-----------
|
|
# Extract numerical data only
|
|
#-----------
|
|
print('\nSelecting numerical data only')
|
|
aaindex_df = aaindex_df.select_dtypes(include = num_type)
|
|
|
|
#---------------------------
|
|
# aaindex: sanity check 1
|
|
#---------------------------
|
|
if len(aaindex_df.columns) == expected_aa_ncols:
|
|
print('\nPASS: successfully selected numerical columns only for aaindex_df')
|
|
else:
|
|
print('\nFAIL: Numbers mismatch'
|
|
, '\nExpected ncols:', expected_aa_ncols
|
|
, '\nGot:', len(aaindex_df.columns))
|
|
|
|
#---------------
|
|
# check for NA
|
|
#---------------
|
|
print('\nNow checking for NA in the remaining aaindex_cols')
|
|
c1 = aaindex_df.isna().sum()
|
|
c2 = c1.sort_values(ascending=False)
|
|
print('\nCounting aaindex_df cols with NA'
|
|
, '\nncols with NA:', sum(c2>0), 'columns'
|
|
, '\nDropping these...'
|
|
, '\nOriginal ncols:', len(aaindex_df.columns)
|
|
)
|
|
aa_df = aaindex_df.dropna(axis=1)
|
|
|
|
print('\nRevised df ncols:', len(aa_df.columns))
|
|
|
|
c3 = aa_df.isna().sum()
|
|
c4 = c3.sort_values(ascending=False)
|
|
|
|
print('\nChecking NA in revised df...')
|
|
|
|
if sum(c4>0):
|
|
sys.exit('\nFAIL: aaindex_df still contains cols with NA, please check and drop these before proceeding...')
|
|
else:
|
|
print('\nPASS: cols with NA successfully dropped from aaindex_df'
|
|
, '\nProceeding with combining aa_df with other features_df')
|
|
|
|
#---------------------------
|
|
# aaindex: sanity check 2
|
|
#---------------------------
|
|
expected_aa_ncols2 = len(aaindex_df.columns) - sum(c2>0)
|
|
if len(aa_df.columns) == expected_aa_ncols2:
|
|
print('\nPASS: ncols match'
|
|
, '\nExpected ncols:', expected_aa_ncols2
|
|
, '\nGot:', len(aa_df.columns))
|
|
else:
|
|
print('\nFAIL: Numbers mismatch'
|
|
, '\nExpected ncols:', expected_aa_ncols2
|
|
, '\nGot:', len(aa_df.columns))
|
|
|
|
# Important: need this to identify aaindex cols
|
|
aa_df_cols = aa_df.columns
|
|
print('\nTotal no. of columns in clean aa_df:', len(aa_df_cols))
|
|
|
|
###############################################################################
|
|
#%% Combining my_features_df and aaindex_df
|
|
#===========================
|
|
# Merge my_df + aaindex_df
|
|
#===========================
|
|
|
|
if aa_df.columns[aa_df.columns.isin(my_features_df.columns)] == my_features_df.columns[my_features_df.columns.isin(aa_df.columns)]:
|
|
print('\nMerging on column: mutationinformation')
|
|
|
|
if len(my_features_df) == len(aa_df):
|
|
expected_nrows = len(my_features_df)
|
|
print('\nProceeding to merge, expected nrows in merged_df:', expected_nrows)
|
|
else:
|
|
sys.exit('\nNrows mismatch, cannot merge. Please check'
|
|
, '\nnrows my_df:', len(my_features_df)
|
|
, '\nnrows aa_df:', len(aa_df))
|
|
|
|
#-----------------
|
|
# Reset index: mutationinformation
|
|
# Very important for merging
|
|
#-----------------
|
|
aa_df = aa_df.reset_index()
|
|
|
|
expected_ncols = len(my_features_df.columns) + len(aa_df.columns) - 1 # for the no. of merging col
|
|
|
|
#-----------------
|
|
# Merge: my_features_df + aa_df
|
|
#-----------------
|
|
merged_df = pd.merge(my_features_df
|
|
, aa_df
|
|
, on = 'mutationinformation')
|
|
|
|
#---------------------------
|
|
# aaindex: sanity check 3
|
|
#---------------------------
|
|
if len(merged_df.columns) == expected_ncols:
|
|
print('\nPASS: my_features_df and aa_df successfully combined'
|
|
, '\nnrows:', len(merged_df)
|
|
, '\nncols:', len(merged_df.columns))
|
|
else:
|
|
sys.exit('\nFAIL: could not combine my_features_df and aa_df'
|
|
, '\nCheck dims and merging cols!')
|
|
|
|
#--------
|
|
# Reassign so downstream code doesn't need to change
|
|
#--------
|
|
my_df = merged_df.copy()
|
|
|
|
#%% Data: my_df
|
|
# Check if non structural pos have crept in
|
|
# IDEALLY remove from source! But for rpoB do it here
|
|
# Drop NA where numerical cols have them
|
|
if gene.lower() in geneL_na_ppi2:
|
|
#D1148 get rid of
|
|
na_index = my_df['mutationinformation'].index[my_df['mcsm_na_affinity'].apply(np.isnan)]
|
|
my_df = my_df.drop(index=na_index)
|
|
|
|
# FIXED: complete data for all muts inc L114M, F115L, V123L, V125I, V131M
|
|
# if gene.lower() in ['embb']:
|
|
# na_index = my_df['mutationinformation'].index[my_df['ligand_distance'].apply(np.isnan)]
|
|
# my_df = my_df.drop(index=na_index)
|
|
|
|
# # Sanity check for non-structural positions
|
|
# print('\nChecking for non-structural postions')
|
|
# na_index = my_df['mutationinformation'].index[my_df['ligand_distance'].apply(np.isnan)]
|
|
# if len(na_index) > 0:
|
|
# print('\nNon-structural positions detected for gene:', gene.lower()
|
|
# , '\nTotal number of these detected:', len(na_index)
|
|
# , '\These are at index:', na_index
|
|
# , '\nOriginal nrows:', len(my_df)
|
|
# , '\nDropping these...')
|
|
# my_df = my_df.drop(index=na_index)
|
|
# print('\nRevised nrows:', len(my_df))
|
|
# else:
|
|
# print('\nNo non-structural positions detected for gene:', gene.lower()
|
|
# , '\nnrows:', len(my_df))
|
|
|
|
|
|
###########################################################################
|
|
#%% Add lineage calculation columns
|
|
#FIXME: Check if this can be imported from config?
|
|
total_mtblineage_uc = 8
|
|
lineage_colnames = ['lineage_list_all', 'lineage_count_all', 'lineage_count_unique', 'lineage_list_unique', 'lineage_multimode']
|
|
#bar = my_df[lineage_colnames]
|
|
my_df['lineage_proportion'] = my_df['lineage_count_unique']/my_df['lineage_count_all']
|
|
my_df['dist_lineage_proportion'] = my_df['lineage_count_unique']/total_mtblineage_uc
|
|
###########################################################################
|
|
#%% Active site annotation column
|
|
# change from numberic to categorical
|
|
|
|
if my_df['active_site'].dtype in num_type:
|
|
my_df['active_site'] = my_df['active_site'].astype(object)
|
|
my_df['active_site'].dtype
|
|
#%% AA property change
|
|
#--------------------
|
|
# Water prop change
|
|
#--------------------
|
|
my_df['water_change'] = my_df['wt_prop_water'] + str('_to_') + my_df['mut_prop_water']
|
|
my_df['water_change'].value_counts()
|
|
|
|
water_prop_changeD = {
|
|
'hydrophobic_to_neutral' : 'change'
|
|
, 'hydrophobic_to_hydrophobic' : 'no_change'
|
|
, 'neutral_to_neutral' : 'no_change'
|
|
, 'neutral_to_hydrophobic' : 'change'
|
|
, 'hydrophobic_to_hydrophilic' : 'change'
|
|
, 'neutral_to_hydrophilic' : 'change'
|
|
, 'hydrophilic_to_neutral' : 'change'
|
|
, 'hydrophilic_to_hydrophobic' : 'change'
|
|
, 'hydrophilic_to_hydrophilic' : 'no_change'
|
|
}
|
|
|
|
my_df['water_change'] = my_df['water_change'].map(water_prop_changeD)
|
|
my_df['water_change'].value_counts()
|
|
|
|
#--------------------
|
|
# Polarity change
|
|
#--------------------
|
|
my_df['polarity_change'] = my_df['wt_prop_polarity'] + str('_to_') + my_df['mut_prop_polarity']
|
|
my_df['polarity_change'].value_counts()
|
|
|
|
polarity_prop_changeD = {
|
|
'non-polar_to_non-polar' : 'no_change'
|
|
, 'non-polar_to_neutral' : 'change'
|
|
, 'neutral_to_non-polar' : 'change'
|
|
, 'neutral_to_neutral' : 'no_change'
|
|
, 'non-polar_to_basic' : 'change'
|
|
, 'acidic_to_neutral' : 'change'
|
|
, 'basic_to_neutral' : 'change'
|
|
, 'non-polar_to_acidic' : 'change'
|
|
, 'neutral_to_basic' : 'change'
|
|
, 'acidic_to_non-polar' : 'change'
|
|
, 'basic_to_non-polar' : 'change'
|
|
, 'neutral_to_acidic' : 'change'
|
|
, 'acidic_to_acidic' : 'no_change'
|
|
, 'basic_to_acidic' : 'change'
|
|
, 'basic_to_basic' : 'no_change'
|
|
, 'acidic_to_basic' : 'change'}
|
|
|
|
my_df['polarity_change'] = my_df['polarity_change'].map(polarity_prop_changeD)
|
|
my_df['polarity_change'].value_counts()
|
|
|
|
#--------------------
|
|
# Electrostatics change
|
|
#--------------------
|
|
my_df['electrostatics_change'] = my_df['wt_calcprop'] + str('_to_') + my_df['mut_calcprop']
|
|
my_df['electrostatics_change'].value_counts()
|
|
|
|
calc_prop_changeD = {
|
|
'non-polar_to_non-polar' : 'no_change'
|
|
, 'non-polar_to_polar' : 'change'
|
|
, 'polar_to_non-polar' : 'change'
|
|
, 'non-polar_to_pos' : 'change'
|
|
, 'neg_to_non-polar' : 'change'
|
|
, 'non-polar_to_neg' : 'change'
|
|
, 'pos_to_polar' : 'change'
|
|
, 'pos_to_non-polar' : 'change'
|
|
, 'polar_to_polar' : 'no_change'
|
|
, 'neg_to_neg' : 'no_change'
|
|
, 'polar_to_neg' : 'change'
|
|
, 'pos_to_neg' : 'change'
|
|
, 'pos_to_pos' : 'no_change'
|
|
, 'polar_to_pos' : 'change'
|
|
, 'neg_to_polar' : 'change'
|
|
, 'neg_to_pos' : 'change'
|
|
}
|
|
|
|
my_df['electrostatics_change'] = my_df['electrostatics_change'].map(calc_prop_changeD)
|
|
my_df['electrostatics_change'].value_counts()
|
|
|
|
#--------------------
|
|
# Summary change: Create a combined column summarising these three cols
|
|
#--------------------
|
|
detect_change = 'change'
|
|
check_prop_cols = ['water_change', 'polarity_change', 'electrostatics_change']
|
|
#my_df['aa_prop_change'] = (my_df.values == detect_change).any(1).astype(int)
|
|
my_df['aa_prop_change'] = (my_df[check_prop_cols].values == detect_change).any(1).astype(int)
|
|
my_df['aa_prop_change'].value_counts()
|
|
my_df['aa_prop_change'].dtype
|
|
|
|
my_df['aa_prop_change'] = my_df['aa_prop_change'].map({1:'change'
|
|
, 0: 'no_change'})
|
|
|
|
my_df['aa_prop_change'].value_counts()
|
|
my_df['aa_prop_change'].dtype
|
|
|
|
#%% IMPUTE values for OR [check script for exploration: UQ_or_imputer]
|
|
#--------------------
|
|
# Impute OR values
|
|
#--------------------
|
|
#or_cols = ['or_mychisq', 'log10_or_mychisq', 'or_fisher']
|
|
sel_cols = ['mutationinformation', 'or_mychisq', 'log10_or_mychisq']
|
|
or_cols = ['or_mychisq', 'log10_or_mychisq']
|
|
|
|
print("count of NULL values before imputation\n")
|
|
print(my_df[or_cols].isnull().sum())
|
|
|
|
my_dfI = pd.DataFrame(index = my_df['mutationinformation'] )
|
|
|
|
|
|
my_dfI = pd.DataFrame(KNN(n_neighbors=3, weights="uniform").fit_transform(my_df[or_cols])
|
|
, index = my_df['mutationinformation']
|
|
, columns = or_cols )
|
|
my_dfI.columns = ['or_rawI', 'logorI']
|
|
my_dfI.columns
|
|
my_dfI = my_dfI.reset_index(drop = False) # prevents old index from being added as a column
|
|
my_dfI.head()
|
|
print("count of NULL values AFTER imputation\n")
|
|
print(my_dfI.isnull().sum())
|
|
|
|
#-------------------------------------------
|
|
# OR df Merge: with original based on index
|
|
#-------------------------------------------
|
|
#my_df['index_bm'] = my_df.index
|
|
mydf_imputed = pd.merge(my_df
|
|
, my_dfI
|
|
, on = 'mutationinformation')
|
|
#mydf_imputed = mydf_imputed.set_index(['index_bm'])
|
|
|
|
my_df['log10_or_mychisq'].isna().sum()
|
|
mydf_imputed['log10_or_mychisq'].isna().sum()
|
|
mydf_imputed['logorI'].isna().sum() # should be 0
|
|
|
|
len(my_df.columns)
|
|
len(mydf_imputed.columns)
|
|
|
|
#-----------------------------------------
|
|
# REASSIGN my_df after imputing OR values
|
|
#-----------------------------------------
|
|
my_df = mydf_imputed.copy()
|
|
|
|
if my_df['logorI'].isna().sum() == 0:
|
|
print('\nPASS: OR values imputed, data ready for ML')
|
|
else:
|
|
sys.exit('\nFAIL: something went wrong, Data not ready for ML. Please check upstream!')
|
|
|
|
#!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!
|
|
#---------------------------------------
|
|
# TODO: try other imputation like MICE
|
|
#---------------------------------------
|
|
#!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!
|
|
|
|
#%%########################################################################
|
|
#==========================
|
|
# Data for ML
|
|
#==========================
|
|
my_df_ml = my_df.copy()
|
|
|
|
#%% Build X: input for ML
|
|
common_cols_stabiltyN = ['ligand_distance'
|
|
, 'ligand_affinity_change'
|
|
, 'duet_stability_change'
|
|
, 'ddg_foldx'
|
|
, 'deepddg'
|
|
, 'ddg_dynamut2'
|
|
, 'mmcsm_lig'
|
|
, 'contacts']
|
|
|
|
# Build stability columns ~ gene
|
|
if gene.lower() in geneL_basic:
|
|
X_stabilityN = common_cols_stabiltyN
|
|
cols_to_mask = ['ligand_affinity_change']
|
|
|
|
if gene.lower() in geneL_ppi2:
|
|
# X_stabilityN = common_cols_stabiltyN + ['mcsm_ppi2_affinity' , 'interface_dist']
|
|
geneL_ppi2_st_cols = ['mcsm_ppi2_affinity', 'interface_dist']
|
|
X_stabilityN = common_cols_stabiltyN + geneL_ppi2_st_cols
|
|
cols_to_mask = ['ligand_affinity_change', 'mcsm_ppi2_affinity']
|
|
|
|
if gene.lower() in geneL_na:
|
|
# X_stabilityN = common_cols_stabiltyN + ['mcsm_na_affinity']
|
|
geneL_na_st_cols = ['mcsm_na_affinity']
|
|
X_stabilityN = common_cols_stabiltyN + geneL_na_st_cols
|
|
cols_to_mask = ['ligand_affinity_change', 'mcsm_na_affinity']
|
|
|
|
if gene.lower() in geneL_na_ppi2:
|
|
# X_stabilityN = common_cols_stabiltyN + ['mcsm_na_affinity'] + ['mcsm_ppi2_affinity', 'interface_dist']
|
|
geneL_na_ppi2_st_cols = ['mcsm_na_affinity'] + ['mcsm_ppi2_affinity', 'interface_dist']
|
|
X_stabilityN = common_cols_stabiltyN + geneL_na_ppi2_st_cols
|
|
cols_to_mask = ['ligand_affinity_change', 'mcsm_na_affinity', 'mcsm_ppi2_affinity']
|
|
|
|
|
|
X_foldX_cols = [ 'electro_rr', 'electro_mm', 'electro_sm', 'electro_ss'
|
|
, 'disulfide_rr', 'disulfide_mm', 'disulfide_sm', 'disulfide_ss'
|
|
, 'hbonds_rr', 'hbonds_mm', 'hbonds_sm', 'hbonds_ss'
|
|
, 'partcov_rr', 'partcov_mm', 'partcov_sm', 'partcov_ss'
|
|
, 'vdwclashes_rr', 'vdwclashes_mm', 'vdwclashes_sm', 'vdwclashes_ss'
|
|
, 'volumetric_rr', 'volumetric_mm', 'volumetric_ss'
|
|
]
|
|
|
|
X_str = ['rsa'
|
|
#, 'asa'
|
|
, 'kd_values'
|
|
, 'rd_values']
|
|
|
|
X_ssFN = X_stabilityN + X_str + X_foldX_cols
|
|
|
|
X_evolFN = ['consurf_score'
|
|
, 'snap2_score'
|
|
, 'provean_score']
|
|
|
|
X_genomic_mafor = ['maf'
|
|
, 'logorI'
|
|
# , 'or_rawI'
|
|
# , 'or_mychisq'
|
|
# , 'or_logistic'
|
|
# , 'or_fisher'
|
|
# , 'pval_fisher'
|
|
]
|
|
|
|
X_genomic_linegae = ['lineage_proportion'
|
|
, 'dist_lineage_proportion'
|
|
#, 'lineage' # could be included as a category but it has L2;L4 formatting
|
|
, 'lineage_count_all'
|
|
, 'lineage_count_unique'
|
|
]
|
|
|
|
X_genomicFN = X_genomic_mafor + X_genomic_linegae
|
|
|
|
X_aaindexFN = list(aa_df_cols)
|
|
|
|
print('\nTotal no. of features for aaindex:', len(X_aaindexFN))
|
|
|
|
# numerical feature names
|
|
numerical_FN = X_ssFN + X_evolFN + X_genomicFN + X_aaindexFN
|
|
|
|
# categorical feature names
|
|
categorical_FN = ['ss_class'
|
|
# , 'wt_prop_water'
|
|
# , 'mut_prop_water'
|
|
# , 'wt_prop_polarity'
|
|
# , 'mut_prop_polarity'
|
|
# , 'wt_calcprop'
|
|
# , 'mut_calcprop'
|
|
, 'aa_prop_change'
|
|
, 'electrostatics_change'
|
|
, 'polarity_change'
|
|
, 'water_change'
|
|
, 'drtype_mode_labels' # beware then you can't use it to predict [USED it for uq_v1, not v2]
|
|
, 'active_site' #[didn't use it for uq_v1]
|
|
#, 'gene_name' # will be required for the combined stuff
|
|
]
|
|
|
|
#----------------------------------------------
|
|
# count numerical and categorical features
|
|
#----------------------------------------------
|
|
|
|
print('\nNo. of numerical features:', len(numerical_FN)
|
|
, '\nNo. of categorical features:', len(categorical_FN))
|
|
|
|
#=======================
|
|
# Masking columns:
|
|
# (mCSM-lig, mCSM-NA, mCSM-ppi2) values for lig_dist >10
|
|
#=======================
|
|
# my_df_ml['mutationinformation'][my_df['ligand_distance']>10].value_counts()
|
|
# my_df_ml.groupby('mutationinformation')['ligand_distance'].apply(lambda x: (x>10)).value_counts()
|
|
|
|
# my_df_ml.loc[(my_df_ml['ligand_distance'] > 10), 'ligand_affinity_change'] = 0
|
|
# (my_df_ml['ligand_affinity_change'] == 0).sum()
|
|
|
|
my_df_ml['mutationinformation'][my_df_ml['ligand_distance']>10].value_counts()
|
|
my_df_ml.groupby('mutationinformation')['ligand_distance'].apply(lambda x: (x>10)).value_counts()
|
|
my_df_ml.loc[(my_df_ml['ligand_distance'] > 10), cols_to_mask].value_counts()
|
|
|
|
# mask the mcsm affinity related columns where ligand distance > 10
|
|
my_df_ml.loc[(my_df_ml['ligand_distance'] > 10), cols_to_mask] = 0
|
|
(my_df_ml['ligand_affinity_change'] == 0).sum()
|
|
|
|
mask_check = my_df_ml[['mutationinformation', 'ligand_distance'] + cols_to_mask]
|
|
|
|
# write file for check
|
|
mask_check.sort_values(by = ['ligand_distance'], ascending = True, inplace = True)
|
|
mask_check.to_csv(outdir + 'ml/' + gene.lower() + '_mask_check.csv')
|
|
|
|
#####################################################################
|
|
#================================================================
|
|
# Training and Blind test [COMPLETE data]: 80/20
|
|
|
|
# Use complete data, call the 20% as blind test
|
|
#================================================================
|
|
my_df_ml[drug].isna().sum()
|
|
# blind_test_df = my_df_ml[my_df_ml[drug].isna()]
|
|
# blind_test_df.shape
|
|
|
|
#training_df = my_df_ml[my_df_ml[drug].notna()]
|
|
#training_df.shape
|
|
|
|
training_df = my_df_ml.copy()
|
|
|
|
# Target 1: dst_mode
|
|
training_df[drug].value_counts()
|
|
training_df['dst_mode'].value_counts()
|
|
|
|
####################################################################
|
|
|
|
###############################################################################
|
|
###############################################################################
|
|
# #%% extracting dfs based on numerical, categorical column names
|
|
# #----------------------------------
|
|
# # WITHOUT the target var included
|
|
# #----------------------------------
|
|
# num_df = training_df[numerical_FN]
|
|
# num_df.shape
|
|
|
|
# cat_df = training_df[categorical_FN]
|
|
# cat_df.shape
|
|
|
|
# all_df = training_df[numerical_FN + categorical_FN]
|
|
# all_df.shape
|
|
|
|
# #------------------------------
|
|
# # WITH the target var included:
|
|
# #'wtgt': with target
|
|
# #------------------------------
|
|
# # drug and dst_mode should be the same thing
|
|
# num_df_wtgt = training_df[numerical_FN + ['dst_mode']]
|
|
# num_df_wtgt.shape
|
|
|
|
# cat_df_wtgt = training_df[categorical_FN + ['dst_mode']]
|
|
# cat_df_wtgt.shape
|
|
|
|
# all_df_wtgt = training_df[numerical_FN + categorical_FN + ['dst_mode']]
|
|
# all_df_wtgt.shape
|
|
|
|
#%%########################################################################
|
|
# #============
|
|
# # ML data: OLD
|
|
# #============
|
|
# #------
|
|
# # X: Training and Blind test (BTS)
|
|
# #------
|
|
# X = all_df_wtgt[numerical_FN + categorical_FN] # training data ALL
|
|
# X_bts = blind_test_df[numerical_FN + categorical_FN] # blind test data ALL
|
|
# #X = all_df_wtgt[numerical_FN] # training numerical only
|
|
# #X_bts = blind_test_df[numerical_FN] # blind test data numerical
|
|
|
|
# #------
|
|
# # y
|
|
# #------
|
|
# y = all_df_wtgt['dst_mode'] # training data y
|
|
# y_bts = blind_test_df['dst_mode'] # blind data test y
|
|
|
|
# #X_bts_wt = blind_test_df[numerical_FN + ['dst_mode']]
|
|
|
|
# # Quick check
|
|
# #(X['ligand_affinity_change']==0).sum() == (X['ligand_distance']>10).sum()
|
|
# for i in range(len(cols_to_mask)):
|
|
# ind = i+1
|
|
# print('\nindex:', i, '\nind:', ind)
|
|
# print('\nMask count check:'
|
|
# , (my_df_ml[cols_to_mask[i]]==0).sum() == (my_df_ml['ligand_distance']>10).sum()
|
|
# )
|
|
|
|
# print('Original Data\n', Counter(y)
|
|
# , 'Data dim:', X.shape)
|
|
|
|
###############################################################################
|
|
###############################################################################
|
|
#====================================
|
|
# ML data: Train test split [COMPLETE data]: 80/20
|
|
# with stratification
|
|
# 80% : training_data for CV
|
|
# 20% : blind test
|
|
#=====================================
|
|
|
|
# features: all_df or
|
|
x_features = training_df[numerical_FN + categorical_FN]
|
|
y_target = training_df['dst_mode']
|
|
|
|
# sanity check
|
|
if not 'dst_mode' in x_features.columns:
|
|
print('\nPASS: x_features has no target variable')
|
|
x_ncols = len(x_features.columns)
|
|
print('\nNo. of columns for x_features:', x_ncols)
|
|
# NEED It for scaling law split
|
|
#https://towardsdatascience.com/finally-why-we-use-an-80-20-split-for-training-and-test-data-plus-an-alternative-method-oh-yes-edc77e96295d
|
|
else:
|
|
sys.exit('\nFAIL: x_features has target variable included. FIX it and rerun!')
|
|
|
|
#x_train, x_test, y_train, y_test # traditional var_names
|
|
# so my downstream code doesn't need to change
|
|
X, X_bts, y, y_bts = train_test_split(x_features, y_target
|
|
, test_size = 0.2
|
|
, **rs
|
|
, stratify = y_target)
|
|
yc1 = Counter(y)
|
|
yc1_ratio = yc1[0]/yc1[1]
|
|
|
|
yc2 = Counter(y_bts)
|
|
yc2_ratio = yc2[0]/yc2[1]
|
|
|
|
print('\n-------------------------------------------------------------'
|
|
, '\nSuccessfully split data with stratification [COMPLETE data]: 80/20'
|
|
, '\nInput features data size:', x_features.shape
|
|
, '\nTrain data size:', X.shape
|
|
, '\nTest data size:', X_bts.shape
|
|
, '\ny_train numbers:', yc1
|
|
, '\ny_train ratio:',yc1_ratio
|
|
, '\n'
|
|
, '\ny_test_numbers:', yc2
|
|
, '\ny_test ratio:', yc2_ratio
|
|
, '\n-------------------------------------------------------------'
|
|
)
|
|
##########################################################################
|
|
# Quick check
|
|
#(X['ligand_affinity_change']==0).sum() == (X['ligand_distance']>10).sum()
|
|
for i in range(len(cols_to_mask)):
|
|
ind = i+1
|
|
print('\nindex:', i, '\nind:', ind)
|
|
print('\nMask count check:'
|
|
, (my_df_ml[cols_to_mask[i]]==0).sum() == (my_df_ml['ligand_distance']>10).sum()
|
|
)
|
|
|
|
print('Original Data\n', Counter(y)
|
|
, 'Data dim:', X.shape)
|
|
###########################################################################
|
|
#%%
|
|
###########################################################################
|
|
# RESAMPLING
|
|
###########################################################################
|
|
#------------------------------
|
|
# Simple Random oversampling
|
|
# [Numerical + catgeorical]
|
|
#------------------------------
|
|
oversample = RandomOverSampler(sampling_strategy='minority')
|
|
X_ros, y_ros = oversample.fit_resample(X, y)
|
|
print('\nSimple Random OverSampling\n', Counter(y_ros))
|
|
print(X_ros.shape)
|
|
|
|
#------------------------------
|
|
# Simple Random Undersampling
|
|
# [Numerical + catgeorical]
|
|
#------------------------------
|
|
undersample = RandomUnderSampler(sampling_strategy='majority')
|
|
X_rus, y_rus = undersample.fit_resample(X, y)
|
|
print('\nSimple Random UnderSampling\n', Counter(y_rus))
|
|
print(X_rus.shape)
|
|
|
|
#------------------------------
|
|
# Simple combine ROS and RUS
|
|
# [Numerical + catgeorical]
|
|
#------------------------------
|
|
oversample = RandomOverSampler(sampling_strategy='minority')
|
|
X_ros, y_ros = oversample.fit_resample(X, y)
|
|
undersample = RandomUnderSampler(sampling_strategy='majority')
|
|
X_rouC, y_rouC = undersample.fit_resample(X_ros, y_ros)
|
|
print('\nSimple Combined Over and UnderSampling\n', Counter(y_rouC))
|
|
print(X_rouC.shape)
|
|
|
|
#------------------------------
|
|
# SMOTE_NC: oversampling
|
|
# [numerical + categorical]
|
|
#https://stackoverflow.com/questions/47655813/oversampling-smote-for-binary-and-categorical-data-in-python
|
|
#------------------------------
|
|
# Determine categorical and numerical features
|
|
numerical_ix = X.select_dtypes(include=['int64', 'float64']).columns
|
|
numerical_ix
|
|
num_featuresL = list(numerical_ix)
|
|
numerical_colind = X.columns.get_indexer(list(numerical_ix) )
|
|
numerical_colind
|
|
|
|
categorical_ix = X.select_dtypes(include=['object', 'bool']).columns
|
|
categorical_ix
|
|
categorical_colind = X.columns.get_indexer(list(categorical_ix))
|
|
categorical_colind
|
|
|
|
k_sm = 5 # 5 is default
|
|
sm_nc = SMOTENC(categorical_features=categorical_colind, k_neighbors = k_sm, **rs, **njobs)
|
|
X_smnc, y_smnc = sm_nc.fit_resample(X, y)
|
|
print('\nSMOTE_NC OverSampling\n', Counter(y_smnc))
|
|
print(X_smnc.shape)
|
|
globals().update(locals()) # TROLOLOLOLOLOLS
|
|
#print("i did a horrible hack :-)")
|
|
###############################################################################
|
|
#%% SMOTE RESAMPLING for NUMERICAL ONLY*
|
|
# #------------------------------
|
|
# # SMOTE: Oversampling
|
|
# # [Numerical ONLY]
|
|
# #------------------------------
|
|
# k_sm = 1
|
|
# sm = SMOTE(sampling_strategy = 'auto', k_neighbors = k_sm, **rs)
|
|
# X_sm, y_sm = sm.fit_resample(X, y)
|
|
# print(X_sm.shape)
|
|
# print('\nSMOTE OverSampling\n', Counter(y_sm))
|
|
# y_sm_df = y_sm.to_frame()
|
|
# y_sm_df.value_counts().plot(kind = 'bar')
|
|
|
|
# #------------------------------
|
|
# # SMOTE: Over + Undersampling COMBINED
|
|
# # [Numerical ONLY]
|
|
# #-----------------------------
|
|
# sm_enn = SMOTEENN(enn=EditedNearestNeighbours(sampling_strategy='all', **rs, **njobs ))
|
|
# X_enn, y_enn = sm_enn.fit_resample(X, y)
|
|
# print(X_enn.shape)
|
|
# print('\nSMOTE Over+Under Sampling combined\n', Counter(y_enn))
|
|
|
|
###############################################################################
|
|
# TODO: Find over and undersampling JUST for categorical data
|