arXiv subjects

Victor Batista

Publications and source records attributed to Victor Batista.

4 recordsLinked to original sources

arXiv2025

A Computational Framework for Simulations of Dissipative Non-Adiabatic Dynamics on Hybrid Oscillator-Qubit Quantum Devices

We introduce a computational framework for simulating non-adiabatic vibronic dynamics on circuit quantum electrodynamics (cQED) platforms. Our approach leverages hybrid oscillator-qubit quantum hardware with mid-circuit measurements and resets, enabling the incorporation of environmental effects such as dissipation and dephasing. To demonstrate its capabilities, we simulate energy transfer dynamics in a triad model of photosynthetic chromophores inspired by natural antenna systems. We specifically investigate the role of dissipation during the relaxation dynamics following photoexcitation, where electronic transitions are coupled to the evolution of quantum vibrational modes. Our results indicate that hybrid oscillator-qubit devices, operating with noise levels below the intrinsic dissipation rates of typical molecular antenna systems, can achieve the simulation fidelity required for practical computations on near-term and early fault-tolerant quantum computing platforms.

quant-ph↗

arXiv2024

Kernel-Elastic Autoencoder for Molecular Design

We introduce the Kernel-Elastic Autoencoder (KAE), a self-supervised generative model based on the transformer architecture with enhanced performance for molecular design. KAE is formulated based on two novel loss functions: modified maximum mean discrepancy and weighted reconstruction. KAE addresses the long-standing challenge of achieving valid generation and accurate reconstruction at the same time. KAE achieves remarkable diversity in molecule generation while maintaining near-perfect reconstructions on the independent testing dataset, surpassing previous molecule-generating models. KAE enables conditional generation and allows for decoding based on beam search resulting in state-of-the-art performance in constrained optimizations. Furthermore, KAE can generate molecules conditional to favorable binding affinities in docking applications as confirmed by AutoDock Vina and Glide scores, outperforming all existing candidates from the training dataset. Beyond molecular design, we anticipate KAE could be applied to solve problems by generation in a wide range of applications.

cs.LG↗

arXiv2020

Using Restricted Boltzmann Machines to Model Molecular Geometries

Precise physical descriptions of molecules can be obtained by solving the Schrodinger equation; however, these calculations are intractable and even approximations can be cumbersome. Force fields, which estimate interatomic potentials based on empirical data, are also time-consuming. This paper proposes a new methodology for modeling a set of physical parameters by taking advantage of the restricted Boltzmann machine's fast learning capacity and representational power. By training the machine on ab initio data, we can predict new data in the distribution of molecular configurations matching the ab initio distribution. In this paper we introduce a new RBM based on the Tanh activation function, and conduct a comparison of RBMs with different activation functions, including sigmoid, Gaussian, and (Leaky) ReLU. Finally we demonstrate the ability of Gaussian RBMs to model small molecules such as water and ethane.

physics.chem-ph↗

arXiv2019

Thermodynamics of the S2-to-S3 State Transition of the Oxygen-Evolving Complex of Photosystem II

The room temperature pump-probe X-ray free electron laser (XFEL) measurements used for serial femtosecond crystallography provide remarkable information about the structures of the catalytic (S-state) intermediates of the oxygen-evolution reaction of photosystem II. However, mixed populations of these intermediates and moderate resolution limit the interpretation of the data from current experiments. The S3 XFEL structures show extra density near the OEC that may correspond to a water/hydroxide molecule. However, in the latest structure, this additional oxygen is 2.08 Å from the Oe2 of D1-E189, which is closer than the sum of the van der Waals radii of the two oxygens. Here, we use Boltzmann statistics and Monte Carlo sampling to provide a model for the S2-to-S3 state transition, allowing structural changes and the insertion of an additional water/hydroxide. Based on our model, water/hydroxide addition to the oxygen-evolving complex (OEC) is not thermodynamically favorable in the S2 g = 2 state, but it is in the S2 g = 4.1 redox isomer. Thus, formation of the S3 state starts by a transition from the S2 g = 2 to the S2 g = 4.1 structure. Then, electrostatic interactions support protonation of D1-H190 and deprotonation of the Ca2+-ligated water (W3) with proton loss to the lumen. The W3 hydroxide moves toward Mn4, completing the coordination shell of Mn4 and moving with its oxidation to Mn(IV) in the S3 state. In addition, binding additional hydroxide to Mn1 leads to a conformational change of D1-E189 in the S2 g = 4.1 and S3 structures. In the S3 state in the population of protonated D1-E189 increases.

physics.bio-ph↗