Table 1 Clinical information, energy and geometrical features: description and values.
From: Machine learning based personalized drug response prediction for lung cancer patients
Feature type | Attributes | Description | Discrete/Continuous | Range |
|---|---|---|---|---|
DCI | Age | Patient’s personal information | Discrete | [0–4] |
Sex | Discrete | [0–2] | ||
Smoking history | Discrete | [0–2] | ||
Response status | Discrete | [0–3] | ||
Energy | VDW | Van der Walls energy | Continuous | [− 60 to − 45] |
EEL | Electrostatic interactions | Continuous | [− 23–11] | |
ESURF | No polar component of the solvation energy | Continuous | [− 45 to − 1] | |
EPB | Polar component of solvation free energy | Continuous | [27–40] | |
Geometric | Matching rates | Matched atoms | Discrete | [0, 17] |
Convex atoms | Strength of interaction | Discrete | [0, 43] | |
Connectivity | Connected atoms | Discrete | [0, 23] | |
Euclidean distance | Distance between drug and target | Continuous | [30–39] | |
Hydrogen bonds | Number of hydrogen bonds | Discrete | [775–1650] |
