Interpretation of Compound Activity Predictions from Complex Machine Learning Models Using Local Approximations and Shapley Values.

By Raquel Rodríguez-Pérez, Jürgen Bajorath. Journal of Medicinal Chemistry, 2020

428 citations recorded by the source corpus.

In qualitative or quantitative studies of structure-activity relationships (SARs), machine learning (ML) models are trained to recognize structural patterns that differentiate between active and inactive compounds. Understanding model decisions is challenging but of critical importance to guide compound design. Moreover, the interpretation of ML results provides an additional level of model validation based on expert knowledge. A number of complex ML approaches, especially deep learning (DL) architectures, have distinctive black-box character. Herein, a locally interpretable explanatory method termed SHapley Additive exPlanations (SHAP) is introduced for rationalizing activity predictions of any ML algorithm, regardless of its complexity. Models resulting from random forest (RF), non-linear support vector machine (SVM), and deep neural network (DNN) calculations are interpreted and structural patterns that increase or reduce the predicted probability of activity are identified and mapped onto test compounds. The results indicate that SHAP has high potential for rationalizing predictions of complex ML models.

Subfields: Computational Theory and Mathematics, Materials Chemistry, Molecular Biology

View Paper: https://www.semanticscholar.org/paper/4fdefd6ca91e7fa80873e51f029f596b07eda70c

How researchers see this paper

The researchers whose trusted networks rank it highest, strongest advocate first. Every researcher gets a different order.

In Computational Theory and Mathematics

  1. Junmei Wang
  2. John C. Dearden

In Materials Chemistry

  1. Tingjun Hou

In Molecular Biology

  1. Tingjun Hou
  2. Zhe Wang
  3. Junmei Wang

Citations in the corpus