The ability to penetrate the blood-brain barrier (BBB) is one of the most important characteristics of potential drug molecules. The numerical measure of overcoming the BBB is usually the so-called logBB parameter. There are a large number of experimental methods for evaluating the ability of molecules to penetrate the BBB, but all of them are very time-consuming. Now scientists have high hopes for machine learning (ML). As part of the work on this project, you will be asked to experiment with various ML algorithms (linear models, boosters, neural networks), as well as with various sets of molecular descriptors to find out which physico-chemical characteristics of molecules best correlate with the ability to penetrate the BBB.
Исакова А.М., Стешин И.С., Шкиль Д.О. (науч. рук. Шитяков С.В.) Accessible and reliable machine learning models for the blood-brain barrier permeability assessment // Сборник тезисов докладов конгресса молодых ученых. Электронное издание. – СПб: Университет ИТМО, [2024]. URL: https://kmu.itmo.ru/digests/article/12443