arXiv subjects

Ivan Rivalta

Publications and source records attributed to Ivan Rivalta.

4 recordsLinked to original sources

arXiv2025

The Quantum Ensemble Variational Optimization Algorithm: Applications to Molecular Inverse Design

Designing molecules with optimized properties remains a fundamental challenge due to the intricate relationship between molecular structure and properties. Traditional computational approaches that address the combinatorial number of possible molecular designs become unfeasible as the molecular size increases, suffering from the so-called `curse of dimensionality' problem. Recent advances in quantum computing hardware present new opportunities to address this problem. Here, we introduce the Quantum Ensemble Variational Optimization (QEVO) method for near-term and early fault-tolerant quantum computing platforms. QEVO efficiently maps molecular structures onto an orthonormal basis of Pauli strings and samples from a superposition state generated by a variational ansatz. The ansatz is iteratively optimized to identify molecular candidates with the desired property. Our numerical simulations demonstrate the potential of QEVO in designing drug-like molecules with anticancer properties, employing a shallow quantum circuit that requires only a modest number of qubits. We envision that QEVO could be applied to a wide range of complex problems, offering practical solutions to problems with combinatorial complexity.

quant-ph↗

arXiv2025

Quantum Chemistry Driven Molecular Inverse Design with Data-free Reinforcement Learning

The inverse design of molecules has challenged chemists for decades. In the past years, machine learning and artificial intelligence have emerged as new tools to generate molecules tailoring desired properties, but with the limit of relying on models that are pretrained on large datasets. Here, we present a data-free generative model based on reinforcement learning and quantum mechanics calculations. To improve the generation, our software is based on a five-model reinforcement learning algorithm designed to mimic the syntactic rules of an original ASCII encoding based on the SMILES one, and here reported. The reinforcement learning generator is rewarded by on-the-fly quantum mechanics calculations within a computational routine addressing conformational sampling. We demonstrate that our software successfully generates new molecules with desired properties finding optimal solutions for problems with known solutions and (sub)optimal molecules for unexplored chemical (sub)spaces, jointly showing significant speed-up to a reference baseline.

physics.chem-ph↗

arXiv2020

Simulating Plasmon Resonances of Gold Nanoparticles with Bipyramidal Shapes by Boundary Element Methods

Computational modeling and accurate simulations of localized surface plasmon resonance (LSPR) absorption properties are reported for gold nanobipyramids (GNBs), a class of metal nanoparticle that features highly tunable, geometrydependent optical properties. GNB bicone models with spherical tips performed best in reproducing experimental LSPR spectra while the comparison with other geometrical models provided a fundamental understanding of base shapes and tip effects on the optical properties of GNBs. Our results demonstrated the importance of averaging all geometrical parameters determined from transmission electron microscopy images to build representative models of GNBs. By assessing the performances of LSPR absorption spectra simulations based on a quasi-static approximation, we provided an applicability range of this approach as a function of the nanoparticle size, paving the way to the theoretical study of the coupling between molecular electron densities and metal nanoparticles in GNB-based nanohybrid systems, with potential applications in the design of nanomaterials for bioimaging, optics and photocatalysis.

physics.comp-ph↗

arXiv2018

Eigenvector Centrality Distribution for Characterization of Protein Allosteric Pathways

Determining the principal energy pathways for allosteric communication in biomolecules, that occur as a result of thermal motion, remains challenging due to the intrinsic complexity of the systems involved. Graph theory provides an approach for making sense of such complexity, where allosteric proteins can be represented as networks of amino acids. In this work, we establish the eigenvector centrality metric in terms of the mutual information, as a mean of elucidating the allosteric mechanism that regulates the enzymatic activity of proteins. Moreover, we propose a strategy to characterize the range of the physical interactions that underlie the allosteric process. In particular, the well known enzyme, imidazol glycerol phosphate synthase (IGPS), is utilized to test the proposed methodology. The eigenvector centrality measurement successfully describes the allosteric pathways of IGPS, and allows to pinpoint key amino acids in terms of their relevance in the momentum transfer process. The resulting insight can be utilized for refining the control of IGPS activity, widening the scope for its engineering. Furthermore, we propose a new centrality metric quantifying the relevance of the surroundings of each residue. In addition, the proposed technique is validated against experimental solution NMR measurements yielding fully consistent results. Overall, the methodologies proposed in the present work constitute a powerful and cost effective strategy to gain insight on the allosteric mechanism of proteins.

q-bio.BM↗