tanu/LSHTM_analysis
1
0
Fork 0
Code Issues Pull requests Releases Wiki Activity

various edits

Browse source
This commit is contained in:
Tanushree Tunstall 2022-07-28 13:19:30 +01:00
parent 90b9477520
commit 8d8a61675f
3 changed files with 601 additions and 1 deletions
Download patch file Download diff file Expand all files Collapse all files

41
scripts/ml/ml_functions/SANITY_CHECK_mask.py Normal file
View file

@ -0,0 +1,41 @@
#!/usr/bin/env python3
# -*- coding: utf-8 -*-
"""
Created on Wed Jul 27 12:17:35 2022
@author: tanu
"""
foo = df[['ligand_distance', 'interface_dist', 'ligand_affinity_change','mcsm_ppi2_affinity']]
cols_to_mask = ['ligand_affinity_change']
cols_to_mask_ppi2 = ['mcsm_ppi2_affinity']
(foo[cols_to_mask+cols_to_mask_ppi2] == 0).sum()
(foo[cols_to_mask+cols_to_mask_ppi2] > 0).sum()
foo.loc[(my_df_ml['ligand_distance'] > 10), cols_to_mask] = 0
infile_ml1 = '/home/tanu/git/Data/ethambutol/output/embb_merged_df3.csv'
bar = pd.read_csv(infile_ml1, index_col = 0)
bar = bar[['ligand_distance', 'interface_dist', 'ligand_affinity_change','mcsm_ppi2_affinity']]
#(bar[cols_to_mask+cols_to_mask_ppi2] == 0).sum()
bar2 = bar.copy()
bar2.loc[(bar2['ligand_distance'] >10), cols_to_mask].value_counts()
bar2.loc[(bar2['ligand_affinity_change'] == 0)].value_counts()
# now change
bar2.loc[(bar2['ligand_distance'] > 10), cols_to_mask] = 0
bar2.loc[(bar2['ligand_affinity_change'] == 0)].value_counts()
bar2.loc[(bar2['ligand_distance'] == 0), cols_to_mask].value_counts()
bar2.loc[(bar2['ligand_distance'] > 10), cols_to_mask].value_counts()
(bar2[cols_to_mask] == 0).sum()
bar2.loc[(bar2['interface_dist'] > 10), cols_to_mask_ppi2] = 0
bar2.loc[(bar2['interface_dist'] > 10), cols_to_mask_ppi2].value_counts()
bar2.loc[(bar2['interface_dist'] == 0), cols_to_mask_ppi2].value_counts()
(bar2[cols_to_mask_ppi2] == 0).sum()
['interface_dist'] + cols_to_mask_ppi2