Seminarium wydziałowe – prelegent: prof. Dariusz Plewczyński (Wydział Matematyki i Nauk Informacyjnych PW) NEW
Zapraszamy na seminarium wydziałowe pt.
Selected Applications of Deep Learning in Chemistry and Molecular Biology
Temat przedstawi prof. Dariusz Plewczyński z Wydziału Matematyki i Nauk Informacyjnych PW.
Termin
9.05.2025 (piątek), godz. 12:15
Miejsce
Gmach Technologii Chemicznej
Audytorium im. prof. Czochralskiego
Abstract
We will start with the description of the current landscape of Deep Learning in chemistry, with real-life examples. We will briefly present some common use of Deep Learning in chemistry, showing notable results and possible, yet unexplored use-cases. As a concrete examples, we will discuss retrosynthesis, predicting the molecules topology, and spectral properties.
We will explore some examples of Deep Learning application for Molecular Design, which is revolutionizing protein and molecular design by enabling machine learning models to operate directly on the non-Euclidean surfaces of proteins. Leveraging the vast structural data available, these methods allow for accurate prediction of docking poses, functional sites in protein-ligand complexes, and the classification of binding pockets and surface motifs. As biology increasingly shifts toward "black box" data paradigms, deep learning approaches offer a path to uncovering interpretable, functional insights critical for drug discovery and biomedical engineering.
Finally, we will review the current state of the knowledge about three-dimensional structure of human genome, focusing on the molecular simulations and Artificial Intelligence models trained on experimental next-generation sequencing and the high resolution microscopy genomic data. We will address the challenges of both static and dynamical chromatin models, with some examples of successful fusion of AI and biophysical polymer simulations proposed recently by our Laboratory.
Udział w seminariach wydziałowych jest obowiązkowy dla doktorantów i dyplomantów.