|
|
Quantitative structure-activity relationship and machine learning
Nierostek, Jakub ; Uhlík, Filip (advisor) ; Svozil, Daniel (referee)
Quantitative structure-activity relationship (QSAR) computational methods allow us to examine the relationship between the chemical structure of molecules and their chemical or biological properties. For QSAR calculations, widely used machine learning methods, such as deep learning models, can be used. In this work, we construct a pipeline for training QSAR machine-learning models that can predict molecular toxicity. Furthermore, we investigate the effect of molecular representation on model performance. Both our deep learning mod- els and traditional machine learning models are employed on Tox21 and Ames Mutagenicity datasets. Their performance is evaluated against recently published models for toxicity prediction using the AUC-ROC metric and, regarding certain toxicity targets, shows improvement over these models. Keywords: QSAR, machine learning, deep learning, molecular descriptors 1
|
|
|
Computer study of protein folding using simplified models
Nierostek, Jakub ; Uhlík, Filip (advisor) ; Jirsák, Jan (referee)
The aim of this work is to design and describe a suitable coarse-grained protein model, on the basis of which protein-folding will be studied. The model will be implemented as a computer program, development of the model in time will be simulated by Hamiltonian Monte Carlo. Using computer simulations, not only the protein-folding itself will be investigated, but also the quantities that characterize the process and the similarity of the real and simulated protein's native conformation. Keywords: protein folding, computer simulation, Hamiltonian Monte Carlo, coarse-grained model 1
|
guest ::
