Graph-Based Feature Selection Approach for Molecular Activity Prediction

Abstract

In the construction of QSAR models for the prediction of molecular activity, feature selection is a common task aimed at improving the results and understanding of the problem. The selection of features allows elimination of irrelevant and redundant features, reduces the effect of dimensionality problems, and improves the generalization and interpretability of the models. In many feature selection applications, such as those based on ensembles of feature selectors, it is necessary to combine different selection processes. In this work, we evaluate the application of a new feature selection approach to the prediction of molecular activity, based on the construction of an undirected graph to combine base feature selectors. The experimental results demonstrate the efficiency of the graph-based method in terms of the classification performance, reduction, and redundancy compared to the standard voting method. The graph-based method can be extended to different feature selection algorithms and applied to other cheminformatics problems.

Supramolecular Chemistry: Exploring the Use of Electronic Structure, Molecular Dynamics, and Machine Learning Approaches

“…The study shows that reducing non‐local interatomic features not only leads to linear computational scaling with system size but also improves the accuracy of the methodology. This has been observed in other cases where feature reduction was used [91–94] . This demonstrates that global features may contain non‐essential elements that can degrade model performance if not pruned.…”

“…This has been observed in other cases where feature reduction was used. [91][92][93][94] This demonstrates that global features may contain non-essential elements that can degrade model performance if not pruned. In the case of a supramolecular buckyball catcher, the authors demonstrate how their approach improves the computational process without compromising the accuracy of the simulations (Figure 16), highlighting the critical balance between reducing feature dimensionality and maintaining essential interaction data for accurate modeling.…”