Utilizing Machine Learning Algorithms for Accurate Prediction of Nanomat erials Cytotoxicity
Valeriano A.P. Bondaug F.G. Ebardo I.P. Banda M.H.T. Almonte P.P. Sabugaa M.A.P. Bagnol J.R.V. Latayada M.J.R. Macalalag J.M.R. Paradero B.D. Mayes M.L. Balanay M.P. Alguno A.C. Capangpangan R.Y.
30 September 2025AMG Transcend Association
Letters in Applied NanoBioScience
2025#14Issue 3
Nanomaterials (NMs) have been widely used in various sectors in recent years, which has prompted efforts to solve the difficulties in synthesizing safe-by-design NMs. The toxicity of NMs can be investigated by in silico approaches, which employ machine learning (ML) algorithms to develop quantitative structure-activity relationship (QSAR) models. If built correctly, these models can predict NMs toxicity with the highest accuracy and reliability. As such, this study aimed to develop several QSAR models that predict the cytotoxicity level of various engineered NMs. Partial least squares discriminant analysis (PLS-DA), random forest (RF), decision trees (DT), and k-nearest neighbors (kNN) were used to develop the QSAR models. Consequently, during internal and external validation, the models achieved F1 scores ranging from 91.03 ̶ 95.23% and 90.52 ̶ 95.19%. Additionally, all models identified exposure time and concentration as highly influential descriptors. Moreover, data clustering based on the most significant descriptors further enhanced the performance of the models, particularly when clustered based on the condition “concentration < 31 μg/ml”. As a result, all models achieved slightly over 95% F1 scores upon internal and external validation. These results imply that the developed QSAR models are highly accurate and reliable.
clustering , cytotoxicity , in silico , machine learning , nanomaterials , QSAR
Text of the article Перейти на текст статьи
Research on Environment and Nanotechnology Laboratories, Research Division, Mindanao State University at Naawan, Misamis Oriental, Naawan, 9023, Philippines
Department of Science and Technology-Science Education Institute, Metro Manila, Taguig City, 1631, Philippines
Department of Mathematics and Statistics, University of Southeastern Philippines, Davao del Sur, Davao City, 8000, Philippines
Department of Mathematics, Caraga State University, Agusan del Norte, Butuan City, 8600, Philippines
Information, Communication and Technology Center, Mindanao State University at Naawan, Misamis Oriental, Naawan, 9023, Philippines
Department of Chemistry and Biochemistry, University of Massachusetts, Dartmouth, 02748, MA, United States
Department of Chemistry, Nazarbayev University, Astana, Kazakhstan
Department of Physics, Mindanao State University-Iligan Institute of Technology, Lanao del Norte, Iligan City, 9200, Philippines
Department of Physical Sciences and Mathematics, Mindanao State University at Naawan, Misamis Oriental, Naawan, 9023, Philippines
Research on Environment and Nanotechnology Laboratories
Department of Science and Technology-Science Education Institute
Department of Mathematics and Statistics
Department of Mathematics
Information
Department of Chemistry and Biochemistry
Department of Chemistry
Department of Physics
Department of Physical Sciences and Mathematics
10 лет помогаем публиковать статьи Международный издатель
Книга Публикация научной статьи Волощук 2026 Book Publication of a scientific article 2026