Encoding prior knowledge in ensemble refinement DOI Creative Commons
Jürgen Köfinger, Gerhard Hummer

Опубликована: Фев. 20, 2024

The proper balancing of information from experiment and theory is a long-standing problem in the analysis noisy incomplete data. Viewed as Pareto optimization problem, improved agreement with experimental data comes at expense growing inconsistencies theoretical reference model. Here, we propose how to set exchange rate priori properly balance this trade-off. We focus on gentle ensemble refinement, where difference between potential energy surfaces refined models small thermal scale. By relating variance Kullback-Leibler divergence respective Boltzmann distributions, one can encode prior knowledge about uncertainties, i.e., force-field errors, rate. uncertainty defined space observables depends their type number, thermodynamic state. highlight relation refinement free perturbation theory. A balanced encoding increases quality transparency refinement. Our findings extend non-Boltzmann becomes an information.

Язык: Английский

Coarse-Grained Model of Disordered RNA for Simulations of Biomolecular Condensates DOI
Ikki Yasuda, Sören von Bülow, Giulio Tesei

и другие.

Journal of Chemical Theory and Computation, Год журнала: 2025, Номер unknown

Опубликована: Фев. 26, 2025

Protein–RNA condensates are involved in a range of cellular activities. Coarse-grained molecular models intrinsically disordered proteins have been developed to shed light on and predict single-chain properties phase separation. An RNA model compatible with such for would enable the study complex biomolecular mixtures involving RNA. Here, we present sequence-independent coarse-grained, two-beads-per-nucleotide disordered, flexible based hydropathy scale. We parametrize model, which term CALVADOS-RNA, using combination bottom-up top-down approaches reproduce local geometry intramolecular interactions atomistic simulations vitro experiments. The semiquantitatively captures several aspects RNA–RNA RNA–protein interactions. examined by comparing calculated experimental virial coefficients nonspecific interaction studying reentrant behavior protein–RNA mixtures. demonstrate utility simulating formation mixed consisting region MED1 chains selective partitioning regions from transcription factors into these compare results Despite simplicity our show that it key may therefore be used as baseline biophysics biology condensates.

Язык: Английский

Процитировано

1
Encoding prior knowledge in ensemble refinement DOI Creative Commons
Jürgen Köfinger, Gerhard Hummer

The Journal of Chemical Physics, Год журнала: 2024, Номер 160(11)

Опубликована: Март 21, 2024

The proper balancing of information from experiment and theory is a long-standing problem in the analysis noisy incomplete data. Viewed as Pareto optimization problem, improved agreement with experimental data comes at expense growing inconsistencies theoretical reference model. Here, we propose how to set exchange rate priori properly balance this trade-off. We focus on gentle ensemble refinement, where difference between potential energy surfaces refined models small thermal scale. By relating variance Kullback-Leibler divergence respective Boltzmann distributions, one can encode prior knowledge about uncertainties, i.e., force-field errors, rate. uncertainty defined space observables depends their type number thermodynamic state. highlight relation refinement free perturbation theory. A balanced encoding increases quality transparency refinement. Our findings extend non-Boltzmann becomes an information.

Язык: Английский

Процитировано

3
Nanosecond chain dynamics of single-stranded nucleic acids DOI Creative Commons
Mark F. Nüesch, Lisa M. Pietrek, Erik D. Holmstrom

и другие.

Nature Communications, Год журнала: 2024, Номер 15(1)

Опубликована: Июль 17, 2024

Abstract The conformational dynamics of single-stranded nucleic acids are fundamental for acid folding and function. However, their elementary chain have been difficult to resolve experimentally. Here we employ a combination single-molecule Förster resonance energy transfer, nanosecond fluorescence correlation spectroscopy, nanophotonic enhancement determine the ensembles rapid short in solution. To interpret experimental results terms end-to-end distance dynamics, utilize hierarchical growth approach, simple polymer models, refinement with Bayesian inference generate structural that closely align data. resulting reconfiguration times exceedingly rapid, 10-ns range. Solvent viscosity-dependent measurements indicate these exhibit negligible internal friction thus dominated by solvent friction. Our provide detailed view distributions acids.

Язык: Английский

Процитировано

2
Simulation-Guided Conformational Space Exploration to Assess Reactive Conformations of a Ribozyme DOI

Sélène Forget,

Marie Juillé,

Élise Duboué-Dijon

и другие.

Journal of Chemical Theory and Computation, Год журнала: 2024, Номер 20(14), С. 6263 - 6277

Опубликована: Июль 3, 2024

Self-splicing ribozymes are small ribonucleic acid (RNA) enzymes that catalyze their own cleavage through a transphosphoesterification reaction. While this process is involved in some specific steps of viral RNA replication and splicing, it also importance the context (putative) first autocatalytic RNA-based systems could have preceded emergence modern life. The uncatalyzed phosphoester bond formation thermodynamically very unfavorable, many experimental studies focused on understanding molecular features catalysis these ribozymes. However, chemical reaction paths short-lived not easily characterized by approaches, so simulation approaches appear as an ideal tool to unveil details Here, we focus model hairpin ribozyme. We show identifying relevant initial conformation for reactivity studies, which frequently overlooked mixed quantum-classical predominantly concentrate itself, can be highly challenging. These challenges stem from limitations both available structures (which chemically altered prevent self-cleavage) accuracy force fields, together with necessity comprehensive sampling. dynamics simulations, combined extensive conformational phase space exploration Hamiltonian replica-exchange enable us characterize basins minimal ribozyme ligated state prior self-cleavage. find what usually considered canonical reactive active site geometries hydrogen-bond patterns optimal addition-elimination general acid/general base metastable only marginally populated. stable appears consistent expectations mechanism does require direct participation residues observations may suffer forcefield inaccuracies, all investigated forcefields lead same conclusions upon proper sampling, contrasting previous investigations shorter timescales suggesting at least one reparametrization Amber99 allowed stabilize aligned conformations. Our study demonstrates most pertinent reactant holds equal alongside accurate determination thermodynamics kinetics

Язык: Английский

Процитировано

1
Encoding prior knowledge in ensemble refinement DOI Creative Commons
Jürgen Köfinger, Gerhard Hummer

Опубликована: Фев. 19, 2024

The proper balancing of information from experiment and theory is a long-standing problem in the analysis noisy incomplete data. Viewed as Pareto optimization problem, improved agreement with experimental data comes at expense growing inconsistencies theoretical reference model. Here, we propose how to set exchange rate priori properly balance this trade-off. We focus on gentle ensemble refinement, where difference between potential energy surfaces refined models small thermal scale. By relating variance Kullback-Leibler divergence respective Boltzmann distributions, one can encode prior knowledge about uncertainties, i.e., force-field errors, rate. uncertainty defined space observables depends their type number, thermodynamic state. highlight relation refinement free perturbation theory. A balanced encoding increases quality transparency refinement. Our findings extend non-Boltzmann becomes an information.

Язык: Английский

Процитировано

0
Encoding prior knowledge in ensemble refinement DOI Creative Commons
Jürgen Köfinger, Gerhard Hummer

Опубликована: Фев. 20, 2024

The proper balancing of information from experiment and theory is a long-standing problem in the analysis noisy incomplete data. Viewed as Pareto optimization problem, improved agreement with experimental data comes at expense growing inconsistencies theoretical reference model. Here, we propose how to set exchange rate priori properly balance this trade-off. We focus on gentle ensemble refinement, where difference between potential energy surfaces refined models small thermal scale. By relating variance Kullback-Leibler divergence respective Boltzmann distributions, one can encode prior knowledge about uncertainties, i.e., force-field errors, rate. uncertainty defined space observables depends their type number, thermodynamic state. highlight relation refinement free perturbation theory. A balanced encoding increases quality transparency refinement. Our findings extend non-Boltzmann becomes an information.

Язык: Английский

Процитировано

0