Anikó Udvarhelyi
Current Organizations / Institutes
The organizations/institutes from the scholar's latest article
Anikó Udvarhelyi has a total of 53 co-authors. Here are the top co-authors by number of shared publications:
- Andreas Menzel
Paul Scherrer Institut, Villigen, Switzerland
3 shared publications - Rainer Wilcken
Novartis Biomedical Research, Basel, Switzerland
3 shared publications - Andreas Winkler
Institute of Biochemistry Graz University of Technology Graz Austria
2 shared publications - Ilme Schlichting
Max Planck Institute for Medical Research, Jahnstrasse 29, Heidelberg, 69120, Germany
2 shared publications Tatiana Domratcheva
Department of Biomolecular Mechanisms, Max-Planck-Institut für Medizinische Forschung , 69120 Heidelberg, Germany
2 shared publications- Tatiana Domratcheva
Department of Chemistry
2 shared publications
Latest Publications
- Open Access
- 15 August 2025
- Published by American Chemical Society (ACS) in Journal of Chemical Information and Modeling
- Vol. 65 (16) , 8624-8636
Quantum Descriptor-Based Machine-Learning Modeling of Thermal Hazard of Cyclic Sulfamidates
Cyclic sulfamidates are commonly used building blocks in organic synthesis. Correct classification of their thermal criticality is crucial for the safe use of these compounds in process development and scale-up. In this study, building on our earlier work (Ferrari et al., 2022), we focused on modeling the reaction enthalpy of a family of 5-membered cyclic sulfamidates toward strong bases. The key challenge for the modeling task was the sparse availability of measured reaction enthalpies, with only 29 measurements available. To address this challenge, we used descriptors based on the quantum-chemical properties of the molecules, as they are more closely related to reaction enthalpies than typical cheminformatics-based descriptors. This approach allowed us to avoid relying solely on data-to-fit models and to focus instead on modeling reaction enthalpies using chemistry-aware techniques, which are more appropriate for small data sets. Three models were constructed using the quantum-chemical descriptors: the first one combining Partial Least Squares (PLS) regression with a Genetic Algorithm (GA), the second one based on the Least Absolute Shrinkage and Selection Operator (LASSO) method, and last, a Gaussian Process Regression (GPR) model. The three models achieved coefficients of determination of 0.78, 0.67, and 0.74, respectively. Although the absolute prediction error values were close to 100 J/g, it is noteworthy that all three techniques provided similar results and accurately classified nearly all compounds into their respective thermal criticality classes. This highlights the methodology's effectiveness in providing a reliable framework for preliminary safety assessment and decision-making in process development.
- 15 August 2025
- Published by American Chemical Society (ACS) in Journal of Chemical Information and Modeling
- Vol. 65 (16) , 8624-8636
Open AccessQuantum Descriptor-Based Machine-Learning Modeling of Thermal Hazard of Cyclic Sulfamidates
Cyclic sulfamidates are commonly used building blocks in organic synthesis. Correct classification of their thermal criticality is crucial for the safe use of these compounds in process development and scale-up. In this study, building on our earlier work (Ferrari et al., 2022), we focused on modeling the reaction enthalpy of a family of 5-membered cyclic sulfamidates toward strong bases. The key challenge for the modeling task was the sparse availability of measured reaction enthalpies, with only 29 measurements available. To address this challenge, we used descriptors based on the quantum-chemical properties of the molecules, as they are more closely related to reaction enthalpies than typical cheminformatics-based descriptors. This approach allowed us to avoid relying solely on data-to-fit models and to focus instead on modeling reaction enthalpies using chemistry-aware techniques, which are more appropriate for small data sets. Three models were constructed using the quantum-chemical descriptors: the first one combining Partial Least Squares (PLS) regression with a Genetic Algorithm (GA), the second one based on the Least Absolute Shrinkage and Selection Operator (LASSO) method, and last, a Gaussian Process Regression (GPR) model. The three models achieved coefficients of determination of 0.78, 0.67, and 0.74, respectively. Although the absolute prediction error values were close to 100 J/g, it is noteworthy that all three techniques provided similar results and accurately classified nearly all compounds into their respective thermal criticality classes. This highlights the methodology's effectiveness in providing a reliable framework for preliminary safety assessment and decision-making in process development.
- 17 August 2020
- Published by Springer Nature in Journal of Computer-Aided Molecular Design
- Vol. 35 (4) , 399-415
ReSCoSS: a flexible quantum chemistry workflow identifying relevant solution conformers of drug-like molecules
Conformational equilibria are at the heart of drug design, yet their energetic description is often hampered by the insufficient accuracy of low-cost methods. Here we present a flexible and semi-automatic workflow based on quantum chemistry, ReSCoSS, designed to identify relevant conformers and predict their equilibria across different solvent environments in the Conductor-like Screening Model for Real Solvents (COSMO-RS) framework. We demonstrate the utility and accuracy of the workflow through conformational case studies on several drug-like molecules from literature where relevant conformations are known. We further show that including ReSCoSS conformers significantly improves COSMO-RS based predictions of physicochemical properties over single-conformation approaches. ReSCoSS has found broad adoption in the in-house drug discovery and development work streams and has contributed to establishing quantum-chemistry methods as a strategic pillar in ligand discovery. Graphic abstract
- 17 August 2020
- Published by Springer Nature in Journal of Computer-Aided Molecular Design
- Vol. 35 (4) , 399-415
ReSCoSS: a flexible quantum chemistry workflow identifying relevant solution conformers of drug-like molecules
Conformational equilibria are at the heart of drug design, yet their energetic description is often hampered by the insufficient accuracy of low-cost methods. Here we present a flexible and semi-automatic workflow based on quantum chemistry, ReSCoSS, designed to identify relevant conformers and predict their equilibria across different solvent environments in the Conductor-like Screening Model for Real Solvents (COSMO-RS) framework. We demonstrate the utility and accuracy of the workflow through conformational case studies on several drug-like molecules from literature where relevant conformations are known. We further show that including ReSCoSS conformers significantly improves COSMO-RS based predictions of physicochemical properties over single-conformation approaches. ReSCoSS has found broad adoption in the in-house drug discovery and development work streams and has contributed to establishing quantum-chemistry methods as a strategic pillar in ligand discovery. Graphic abstract
- 6 November 2019
- Published by American Chemical Society (ACS) in Journal of Chemical Information and Modeling
- Vol. 59 (11) , 4806-4813
Benchmarking Different QM Levels for Usage with COSMO-RS
The COSMO-RS method is an established method for the prediction of fluid phase properties such as activity coefficients, liquid–liquid equilibria, and free energy of solvation. It is also frequently used in quantum chemistry-based chemical reaction modeling to predict the solvation contribution to the reactions. The COSMOtherm software, which features the currently most advanced implementation of COSMO-RS, is based on quantum chemical COSMO calculations using the BP functional with the def2-TZVPD basis set. As the accuracy of COSMO-RS depends on the accuracy of the underlying quantum chemical (QC) calculation, it is important to validate the currently used level against other common, presumably superior, approaches such as the more recently developed M06-2x hybrid density functional or wave function methods such as MP2. As compared to other applications where the electronic energy is the most important result of the QC calculation, the COSMO-RS method has a much higher dependence on the molecular polarity and thus the electron density distribution. We find that MP2, PBE0, and M06-2x perform slightly worse in fully reparametrized COSMO-RS with respect to the prediction of experimentally measured properties like pKa or logP. Although MP2 was reported to yield better polarities than most DFT functionals for spin unpolarized molecules, this theoretical advantage does not manifest in a practical benefit for the prediction of thermodynamic properties with a refitted COSMO-RS parameter set. Other pure DFT functionals such as PBE or TPSS can be used instead of BP, but again, no practical advantage is expected as they yield extremely similar polarities to the original BP calculations.
- 6 November 2019
- Published by American Chemical Society (ACS) in Journal of Chemical Information and Modeling
- Vol. 59 (11) , 4806-4813
Benchmarking Different QM Levels for Usage with COSMO-RS
The COSMO-RS method is an established method for the prediction of fluid phase properties such as activity coefficients, liquid–liquid equilibria, and free energy of solvation. It is also frequently used in quantum chemistry-based chemical reaction modeling to predict the solvation contribution to the reactions. The COSMOtherm software, which features the currently most advanced implementation of COSMO-RS, is based on quantum chemical COSMO calculations using the BP functional with the def2-TZVPD basis set. As the accuracy of COSMO-RS depends on the accuracy of the underlying quantum chemical (QC) calculation, it is important to validate the currently used level against other common, presumably superior, approaches such as the more recently developed M06-2x hybrid density functional or wave function methods such as MP2. As compared to other applications where the electronic energy is the most important result of the QC calculation, the COSMO-RS method has a much higher dependence on the molecular polarity and thus the electron density distribution. We find that MP2, PBE0, and M06-2x perform slightly worse in fully reparametrized COSMO-RS with respect to the prediction of experimentally measured properties like pKa or logP. Although MP2 was reported to yield better polarities than most DFT functionals for spin unpolarized molecules, this theoretical advantage does not manifest in a practical benefit for the prediction of thermodynamic properties with a refitted COSMO-RS parameter set. Other pure DFT functionals such as PBE or TPSS can be used instead of BP, but again, no practical advantage is expected as they yield extremely similar polarities to the original BP calculations.
- 23 August 2018
- Published by Springer Nature in Journal of Computer-Aided Molecular Design
- Vol. 32 (10) , 1139-1149
High accuracy quantum-chemistry-based calculation and blind prediction of macroscopic pKa values in the context of the SAMPL6 challenge
Recent advances in the development of low-cost quantum chemical methods have made the prediction of conformational preferences and physicochemical properties of medium-sized drug-like molecules routinely feasible, with significant potential to advance drug discovery. In the context of the SAMPL6 challenge, macroscopic pKa values were blindly predicted for a set of 24 of such molecules. In this paper we present two similar quantum chemical based approaches based on the high accuracy calculation of standard reaction free energies and the subsequent determination of those pKa values via a linear free energy relationship. Both approaches use extensive conformational sampling and apply hybrid and double-hybrid density functional theory with continuum solvation to calculate free energies. The blindly calculated macroscopic pKa values were in excellent agreement with the experiment.
- 23 August 2018
- Published by Springer Nature in Journal of Computer-Aided Molecular Design
- Vol. 32 (10) , 1139-1149
High accuracy quantum-chemistry-based calculation and blind prediction of macroscopic pKa values in the context of the SAMPL6 challenge
Recent advances in the development of low-cost quantum chemical methods have made the prediction of conformational preferences and physicochemical properties of medium-sized drug-like molecules routinely feasible, with significant potential to advance drug discovery. In the context of the SAMPL6 challenge, macroscopic pKa values were blindly predicted for a set of 24 of such molecules. In this paper we present two similar quantum chemical based approaches based on the high accuracy calculation of standard reaction free energies and the subsequent determination of those pKa values via a linear free energy relationship. Both approaches use extensive conformational sampling and apply hybrid and double-hybrid density functional theory with continuum solvation to calculate free energies. The blindly calculated macroscopic pKa values were in excellent agreement with the experiment.
- 5 April 2016
- Published by American Chemical Society (ACS) in The Journal of Physical Chemistry B
- Vol. 120 (14) , 3493-3502
Onset of the Electronic Absorption Spectra of Isolated and π-Stacked Oligomers of 5,6-Dihydroxyindole: An Ab Initio Study of the Building Blocks of Eumelanin
Eumelanin is a naturally occurring skin pigment which is responsible for developing a suntan. The complex structure of eumelanin consists of π-stacked oligomers of various indole derivatives, such as the monomeric building block 5,6-dihydroxyindole (DHI). In this work, we present an ab initio wave-function study of the absorption behavior of DHI oligomers and of doubly and triply π-stacked species of these oligomers. We have simulated the onset of the electronic absorption spectra by employing the MP2 and the linear-response CC2 methods. Our results demonstrate the effect of an increasing degree of oligomerization of DHI and of an increasing degree of π-stacking of DHI oligomers on the onset of the absorption spectra and on the degree of red-shift toward the visible region of the spectrum. We find that π-stacking of DHI and its oligomers substantially red-shifts the onset of the absorption spectra. Our results also suggest that the optical properties of biological eumelanin cannot be simulated by considering the DHI building blocks alone, but instead the building blocks indole-semiquinone and indole-quinone have to be considered as well. This study contributes to advancing the understanding of the complex photophysics of the eumelanin biopolymer.
- 5 April 2016
- Published by American Chemical Society (ACS) in The Journal of Physical Chemistry B
- Vol. 120 (14) , 3493-3502
Onset of the Electronic Absorption Spectra of Isolated and π-Stacked Oligomers of 5,6-Dihydroxyindole: An Ab Initio Study of the Building Blocks of Eumelanin
Eumelanin is a naturally occurring skin pigment which is responsible for developing a suntan. The complex structure of eumelanin consists of π-stacked oligomers of various indole derivatives, such as the monomeric building block 5,6-dihydroxyindole (DHI). In this work, we present an ab initio wave-function study of the absorption behavior of DHI oligomers and of doubly and triply π-stacked species of these oligomers. We have simulated the onset of the electronic absorption spectra by employing the MP2 and the linear-response CC2 methods. Our results demonstrate the effect of an increasing degree of oligomerization of DHI and of an increasing degree of π-stacking of DHI oligomers on the onset of the absorption spectra and on the degree of red-shift toward the visible region of the spectrum. We find that π-stacking of DHI and its oligomers substantially red-shifts the onset of the absorption spectra. Our results also suggest that the optical properties of biological eumelanin cannot be simulated by considering the DHI building blocks alone, but instead the building blocks indole-semiquinone and indole-quinone have to be considered as well. This study contributes to advancing the understanding of the complex photophysics of the eumelanin biopolymer.
- 20 July 2015
- Published by American Chemical Society (ACS) in Journal of Chemical Theory and Computation
- Vol. 11 (8) , 3878-3894
Role of the Molecular Environment in Flavoprotein Color and Redox Tuning: QM Cluster versus QM/MM Modeling
We investigate the origin of the excitation energy shifts induced by the apoprotein in the active site of the bacterial photoreceptor BLUF (Blue Light sensor Using Flavin adenine dinucleotide). In order to compute the vertical excitation energies of three low-lying electronic states, including two π–π* states of flavin (S1 and S2) and a π–π* tyrosine-flavin electron-transfer state (ET), with respect to the energy of the closed-shell ground state (S0), we prepared alternative quantum mechanical (QM) cluster and quantum mechanics/molecular mechanics (QM/MM) models. We found that the excitation energies computed with both types of models correlate with the magnitude of the charge transfer character of the excitation. Accordingly, we conclude that the small charge transfer character of the light absorbing S0–S1 transition and the substantial charge transfer character of the nonabsorbing but redox active S0–ET transition explain the small color changes but substantial redox tuning in BLUF and also in other flavoproteins. Further analysis showed that redox tuning is governed by the electrostatic interaction in the QM/MM model and transfer of charge between the active site and its environment in the QM cluster. Moreover, the wave function polarization of the QM subsystem by the MM subsystem influences the magnitude of the charge transfer, resulting in the QM/MM and QM excitation energies that are not entirely consistent.
- 20 July 2015
- Published by American Chemical Society (ACS) in Journal of Chemical Theory and Computation
- Vol. 11 (8) , 3878-3894
Role of the Molecular Environment in Flavoprotein Color and Redox Tuning: QM Cluster versus QM/MM Modeling
We investigate the origin of the excitation energy shifts induced by the apoprotein in the active site of the bacterial photoreceptor BLUF (Blue Light sensor Using Flavin adenine dinucleotide). In order to compute the vertical excitation energies of three low-lying electronic states, including two π–π* states of flavin (S1 and S2) and a π–π* tyrosine-flavin electron-transfer state (ET), with respect to the energy of the closed-shell ground state (S0), we prepared alternative quantum mechanical (QM) cluster and quantum mechanics/molecular mechanics (QM/MM) models. We found that the excitation energies computed with both types of models correlate with the magnitude of the charge transfer character of the excitation. Accordingly, we conclude that the small charge transfer character of the light absorbing S0–S1 transition and the substantial charge transfer character of the nonabsorbing but redox active S0–ET transition explain the small color changes but substantial redox tuning in BLUF and also in other flavoproteins. Further analysis showed that redox tuning is governed by the electrostatic interaction in the QM/MM model and transfer of charge between the active site and its environment in the QM cluster. Moreover, the wave function polarization of the QM subsystem by the MM subsystem influences the magnitude of the charge transfer, resulting in the QM/MM and QM excitation energies that are not entirely consistent.
- 17 November 2014
- Published by American Chemical Society (ACS) in ACS Chemical Biology
- Vol. 10 (2) , 502-509
Structural Details of Light Activation of the LOV2-based Photoswitch PA-Rac1
Optical control of cellular processes is an emerging approach for studying biological systems, affording control with high spatial and temporal resolution. Specifically designed artificial photoswitches add an interesting extension to naturally occurring light-regulated functionalities. However, despite a great deal of structural information, the generation of new tools cannot be based fully on rational design yet; in many cases design is limited by our understanding of molecular details of light activation and signal transduction. Our biochemical and biophysical studies on the established optogenetic tool PA-Rac1, the photoactivatable small GTPase Rac1, reveal how unexpected details of the sensor–effector interface, such as metal coordination, significantly affect functionally important structural elements of this photoswitch. Together with solution scattering experiments, our results favor differences in the population of pre-existing conformations as the underlying allosteric activation mechanism of PA-Rac1, rather than the assumed release of the Rac1 domain from the caging photoreceptor domain. These results have implications for the design of new optogenetic tools and highlight the importance of including molecular details of the sensor–effector interface, which is however difficult to assess during the initial design of novel artificial photoswitches.
- 17 November 2014
- Published by American Chemical Society (ACS) in ACS Chemical Biology
- Vol. 10 (2) , 502-509
Structural Details of Light Activation of the LOV2-based Photoswitch PA-Rac1
Optical control of cellular processes is an emerging approach for studying biological systems, affording control with high spatial and temporal resolution. Specifically designed artificial photoswitches add an interesting extension to naturally occurring light-regulated functionalities. However, despite a great deal of structural information, the generation of new tools cannot be based fully on rational design yet; in many cases design is limited by our understanding of molecular details of light activation and signal transduction. Our biochemical and biophysical studies on the established optogenetic tool PA-Rac1, the photoactivatable small GTPase Rac1, reveal how unexpected details of the sensor–effector interface, such as metal coordination, significantly affect functionally important structural elements of this photoswitch. Together with solution scattering experiments, our results favor differences in the population of pre-existing conformations as the underlying allosteric activation mechanism of PA-Rac1, rather than the assumed release of the Rac1 domain from the caging photoreceptor domain. These results have implications for the design of new optogenetic tools and highlight the importance of including molecular details of the sensor–effector interface, which is however difficult to assess during the initial design of novel artificial photoswitches.
- 6 March 2014
- Published by Springer Nature
- Vol. 1146, 191-228
Computational Spectroscopy, Dynamics, and Photochemistry of Photosensory Flavoproteins
Extensive interest in photosensory proteins stimulated computational studies of flavins and flavoproteins in the past decade. This review is dedicated to the three central topics of these studies: calculations of flavin UV–visible and IR spectra, simulated dynamics of photoreceptor proteins, and flavin photochemistry. Accordingly, this chapter is divided into three parts; each part describes corresponding computational protocols, summarizes computational results, and discusses the emerging mechanistic picture.
- 6 March 2014
- Published by Springer Nature
- Vol. 1146, 191-228
Computational Spectroscopy, Dynamics, and Photochemistry of Photosensory Flavoproteins
Extensive interest in photosensory proteins stimulated computational studies of flavins and flavoproteins in the past decade. This review is dedicated to the three central topics of these studies: calculations of flavin UV–visible and IR spectra, simulated dynamics of photoreceptor proteins, and flavin photochemistry. Accordingly, this chapter is divided into three parts; each part describes corresponding computational protocols, summarizes computational results, and discusses the emerging mechanistic picture.
- Open Access
- 26 November 2013
- Published by Elsevier in Journal of Molecular Biology
- Vol. 426 (4) , 853-868
Characterization of Elements Involved in Allosteric Light Regulation of Phosphodiesterase Activity by Comparison of Different Functional BlrP1 States
- 26 November 2013
- Published by Elsevier in Journal of Molecular Biology
- Vol. 426 (4) , 853-868
Open AccessCharacterization of Elements Involved in Allosteric Light Regulation of Phosphodiesterase Activity by Comparison of Different Functional BlrP1 States
- 4 March 2013
- Published by American Chemical Society (ACS) in The Journal of Physical Chemistry B
- Vol. 117 (10) , 2888-2897
Glutamine Rotamers in BLUF Photoreceptors: A Mechanistic Reappraisal
The blue light using FAD (BLUF) photosensory protein domain is activated by a unique photoreaction that results in a hydrogen-bond rearrangement around the flavin chromophore. The chemical structure of the hydrogen bond switch is a long-standing debate: The two main hypotheses postulate rotation as opposed to tautomerization of a conserved glutamine residue. Attempts to resolve the debate were inconclusive so far, despite numerous experimental and computational studies. Here we propose physical criteria for the dark and light state structures as well as for the light-activation process to evaluate existing models of BLUF using quantum-chemical calculations. The glutamine rotamer assignment of the crystal structure with the pdb code 1YRX does not satisfy our criteria because after equilibrating the intermolecular forces the glutamine rotamer in 1YRX is incompatible with the experimental density. We identified the root of the mechanistic controversy in the incorrect glutamine rotamer assignment of 1YRX. Furthermore, we show that the glutamine side chain may rotate without light activation in the BLUF dark state. Finally, we demonstrate that the tautomerized glutamine is consistent with our criteria and observations of the BLUF light state.
- 4 March 2013
- Published by American Chemical Society (ACS) in The Journal of Physical Chemistry B
- Vol. 117 (10) , 2888-2897
Glutamine Rotamers in BLUF Photoreceptors: A Mechanistic Reappraisal
The blue light using FAD (BLUF) photosensory protein domain is activated by a unique photoreaction that results in a hydrogen-bond rearrangement around the flavin chromophore. The chemical structure of the hydrogen bond switch is a long-standing debate: The two main hypotheses postulate rotation as opposed to tautomerization of a conserved glutamine residue. Attempts to resolve the debate were inconclusive so far, despite numerous experimental and computational studies. Here we propose physical criteria for the dark and light state structures as well as for the light-activation process to evaluate existing models of BLUF using quantum-chemical calculations. The glutamine rotamer assignment of the crystal structure with the pdb code 1YRX does not satisfy our criteria because after equilibrating the intermolecular forces the glutamine rotamer in 1YRX is incompatible with the experimental density. We identified the root of the mechanistic controversy in the incorrect glutamine rotamer assignment of 1YRX. Furthermore, we show that the glutamine side chain may rotate without light activation in the BLUF dark state. Finally, we demonstrate that the tautomerized glutamine is consistent with our criteria and observations of the BLUF light state.