Classification models for predicting the antimalarial activity against Plasmodium falciparum

ABSTRACT

Support vector machine (SVM) and general regression neural network (GRNN) were used to develop classification models for predicting the antimalarial activity against Plasmodium falciparum. Only 15 molecular descriptors were used to build the classification models for the antimalarial activities of 4750 compounds, which were divided into a training set (3887 compounds) and a test set (863 compounds). For the SVM model, its prediction accuracies are 89.5% for the training set and 87.3% for the test set. For the GRNN model, the prediction accuracies for the two sets are 99.7% and 88.9%, respectively. Both SVC and GRNN models have better prediction ability than the classification model based on binary logistic regression (BLR) analysis. Compared with previously published classification models both SVC and GRNN models are satisfactory in predicting antimalarial activities of compounds with in addition of fewer descriptors.

KEYWORDS:

• Antimalarial
• binary logistic regression
• classification model
• general regression neural network
• support vector classification

Disclosure statement

No potential conflict of interest was reported by the authors.

Supplementary Material

Supplemental data for this article can be accessed at: https://doi.org/10.1080/1062936X.2020.1740890.

Additional information

Funding

This work was supported by the Program for Innovative Research Team of Huizhou University, the Doctoral Initiative Research Program of Huizhou University [2018JB011]; and the Open Project Program of Hunan Provincial Key Laboratory of Environmental Catalysis & Waste Regeneration (Hunan Institute of Engineering) [2018KF11]. The authors wish to express their sincere thanks to Professor Xinliang Yu at Hunan Institute of Engineering for his guidance.