A5043568004- Synthesis and Properties of Aromatic Compounds
- Advanced Chemical Physics Studies
- Fullerene Chemistry and Applications
- Computational Drug Discovery Methods
- Organic Electronics and Photovoltaics
- Machine Learning in Materials Science
- Analytical Chemistry and Sensors
- Tuberculosis Research and Epidemiology
- Quantum, superfluid, helium dynamics
- Fusion materials and technologies
- Mass Spectrometry Techniques and Applications
- Analytical Chemistry and Chromatography
- Cold Atom Physics and Bose-Einstein Condensates
- Plasma Diagnostics and Applications
- Essential Oils and Antimicrobial Activity
- Spectroscopy and Laser Applications
- nanoparticles nucleation surface interactions
- Analytical chemistry methods development
- Gas Sensing Nanomaterials and Sensors
- Plant biochemistry and biosynthesis
- Laser Design and Applications
- Plasma Applications and Diagnostics
Slovak University of Technology in Bratislava
2019-2024
Institute of Inorganic Chemistry of the Slovak Academy of Sciences
2019
Comenius University Bratislava
2011
A series of novel hetero[n]circulenes the first, second, and third generation including S Se heteroatoms for n = 6, 7, 8, 9 were theoretically designed. Their chemical electronic structures investigated using B3LYP-based computational method. The interaction energies electric drift mobilities evaluated model parallel-stacked parallel-slipped dimer configurations at room temperature Marcus theory Einstein relation. On basis calculated properties, second sunflower molecules containing...
Abstract Here we present three distinct machine learning (ML) approaches (TensorFlow, XGBoost, and SchNetPack) for docking score prediction. AutoDock Vina is used to evaluate the inhibitory potential of ZINC15 in‐vivo in‐vitro‐only sets towards SARS‐CoV‐2 main protease. The set (59 884 compounds) ML training (max. 80%), validation (5%), testing (15%). (174 014 evaluation prediction capability trained models. Contributions error are analyzed with respect compounds' charge, number atoms,...
Abstract Diffusion coefficient depends on temperature, pressure, reduced mass of colliding particles and collision cross section. The presented method is designed to calculate the diffusion in loose systems containing molecules with relatively complicated trajectories. It a combination Chapman-Enskog theory molecular dynamics calculation. provides relation between section which result multiple integration scattering angle all possible initial conditions collision. obtained by numerical...
Options to improve the extrapolation power of neural network designed using SchNetPack package with respect top docking scores prediction are presented. It is shown that hyperparameter tuning atomistic model representation (in schnetpack.representation) improves scoring compounds, which have characteristically a low incidence in randomized data sets for training machine learning models. The robustness evaluated according mean square error (MSE) and entropy average loss landscape decrease....
Abstract A computational study using density functional theory is reported for selected model aza[n]circulenes (n = 6, 7, 8 and 9) their derivatives consisting of pyrrole benzene units. Local aromaticity central rings was discussed analyzed theoretical structural indices. Depending on molecular structures, energies the highest occupied lowest unoccupied orbitals change from –5.23 eV to –4.08 –1.97 –0.41 eV, respectively. Based B3LYP calculated optimal geometries, electronic structure...
CCSD(T) ground state potential curves of Pb⋯RG systems (RG = He, Ne and Ar) are presented the importance inclusion spin–orbit effects is discussed. The closed-shell character Pb atom at two-component relativistic level theory leads to shallower energy compared scalar open-shell calculations. pressure-independent cross-diffusion coefficients pD12 have been simulated using extrapolated curves. diffusion from scattering with simulations based on molecular dynamics (MD) velocity autocorrelation...
Abstract Formation and fragmentation of beryllium clusters up to 100 atoms were studied using molecular dynamics trajectory simulations. Association rate constants a single Be atom attachment Ben clusters, n = 2-20, 30, 40, 50, 60, 70, 80, 90 100, dissociation these calculated directly. Dependence the on temperature is strong follows Arrhenius equation. An analysis reactions has shown that dissociate mainly into atoms. The monomer association reaction depends weakly but strongly size cluster.
Abstract A systematic quantum chemical study of model linear oligomers and their ring-fused (condensed) analogues based on six-membered five-membered aromatic units is presented. Static electric polarizabilities were calculated for optimal geometries. The dependence electronic vibrational polarizability contributions the molecular size was discussed polymer limits estimated. presence a diradical structure in fused compounds significantly increases values.