On April 5, we organized a scientific seminar on “Using Neural Networks for Prediction of Properties of Molecules  “. Marián Gall, a lecturer at Slovak University of Technology in Bratislava, Slovakia, led the seminar, which was organized in the framework of the FrontSeat project as part of the seminar series on “Research Seminar on Smart Cybernetics”.

Recently, the use of neural networks has become increasingly crucial in the field of computational chemistry and drug design for predicting molecular properties. By harnessing the power of neural networks, the research aims to accurately forecast various molecular attributes, facilitating a deeper understanding of molecular behavior and interactions. This innovative approach promises to enhance the efficiency of molecular simulations and property prediction, offering significant advancements in the fields of materials science and pharmaceuticals.

In the second part, the specific study delves into the use of neural networks for predicting molecular properties, crucial in computational chemistry and drug design. It highlights the novel application of neural networks to forecast the inhibitory potential of SARS-CoV-2 main protease molecules, leveraging advanced tools like Keras, TensorFlow, DScribe’s SOAP, and AutoDock Vina. Our approach divides a substantial dataset for training, validation, and testing, introducing a method for prediction error estimation that enhances the accuracy of our machine learning models. This neural network model leads to a significant reduction in the number of compounds requiring investigation, showcasing the potential of our models to streamline the screening process. The results affirm the models’ consistent predictive
capacity of the docking scores for used in-vitro datasets of molecules.

This project has received funding from the European Union’s Horizon under grant no. 101079342 (Fostering Opportunities Towards Slovak Excellence in Advanced Control for Smart Industries).


Leave a Reply

Avatar placeholder

Your email address will not be published. Required fields are marked *