Thirty years of molecular dynamics simulations on posttranslational modifications of proteins

Weigle, Austin T.; Feng, Jiangyan; Shukla, Diwakar

doi:10.1039/d2cp02883b

Cited by 15 publications

(10 citation statements)

References 285 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Molecular dynamics (MD) simulations have become a widely applied computational tool to disentangle the details of nanoscopic systems relevant to a wide range of fields, from materials engineering 1,2 to fundamental biology. 3 The reason for their widespread use lies in their ability to resolve the dynamics of molecular systems at excellent time and space resolutions. Nonetheless, the fine-grained time steps of MD simulations also result in high computational costs if the processes to be observed occur at long time scales.…”

Section: Introductionmentioning

confidence: 99%

Active Learning of the Conformational Ensemble of Proteins Using Maximum Entropy VAMPNets

Kleiman

Shukla

2023

J. Chem. Theory Comput.

Self Cite

View full text Add to dashboard Cite

Rapid computational exploration of the free energy landscape of biological molecules remains an active area of research due to the difficulty of sampling rare state transitions in molecular dynamics (MD) simulations. In recent years, an increasing number of studies have exploited machine learning (ML) models to enhance and analyze MD simulations. Notably, unsupervised models that extract kinetic information from a set of parallel trajectories have been proposed including the variational approach for Markov processes (VAMP), VAMPNets, and time-lagged variational autoencoders (TVAE). In this work, we propose a combination of adaptive sampling with active learning of kinetic models to accelerate the discovery of the conformational landscape of biomolecules. In particular, we introduce and compare several techniques that combine kinetic models with two adaptive sampling regimes (least counts and multiagent reinforcement learningbased adaptive sampling) to enhance the exploration of conformational ensembles without introducing biasing forces. Moreover, inspired by the active learning approach of uncertainty-based sampling, we also present MaxEnt VAMPNet. This technique consists of restarting simulations from the microstates that maximize the Shannon entropy of a VAMPNet trained to perform the soft discretization of metastable states. By running simulations on two test systems, the WLALL pentapeptide and the villin headpiece subdomain, we empirically demonstrate that MaxEnt VAMPNet results in faster exploration of conformational landscapes compared with the baseline and other proposed methods.

show abstract

Section: Introductionmentioning

confidence: 99%

Active Learning of the Conformational Ensemble of Proteins Using Maximum Entropy VAMPNets

Kleiman

Shukla

2023

J. Chem. Theory Comput.

Self Cite

View full text Add to dashboard Cite

show abstract

“…Molecular dynamics (MD) simulations have become a widely-applied computational tool to disentangle the details of nanoscopic systems relevant to a wide range of fields, from materials engineering 1,2 to fundamental biology. 3 The reason for their widespread use lies in their ability to resolve the dynamics of molecular systems at excellent time and space resolutions. Nonetheless, the fine-grained time steps of MD simulations also result in high computational costs if the processes to be observed occur at long timescales.…”

Section: Introductionmentioning

confidence: 99%

Active Learning of the Conformational Ensemble of Proteins using Maximum Entropy VAMPNets

Kleiman

Shukla

2023

Preprint

Self Cite

View full text Add to dashboard Cite

Rapid computational exploration of the free energy landscape of biological molecules remains an active area of research due to the difficulty of sampling rare state transitions in Molecular Dynamics (MD) simulations. In recent years, an increasing number of studies have exploited Machine Learning (ML) models to enhance and analyze MD simulations. Notably, unsupervised models that extract kinetic information from a set of parallel trajectories have been proposed, including the variational approach for Markov processes (VAMP), VAMPNets, and time-lagged variational autoencoders (TVAE). In this work, we propose a combination of adaptive sampling with active learning of kinetic models to accelerate the discovery of the conformational landscape of biomolecules. In particular, we introduce and compare several techniques that combine kinetic models with two adaptive sampling regimes (least counts and multi-agent reinforcement learning-based adaptive sampling) to enhance the exploration of conformational ensembles without introducing biasing forces. Moreover, inspired by the active learning approach of uncertainty-based sampling, we also present MaxEnt VAMPNet. This technique consists of restarting simulations from the microstates that maximize the Shannon entropy of a VAMPNet trained to perform soft discretization of metastable states. By running simulations on two test systems, the WLALL pentapeptide and the villin headpiece subdomain, we empirically demonstrate that MaxEnt VAMPNet results in faster exploration of conformational landscapes compared to the baseline and other proposed methods.

show abstract

“…Glutathione transferase enzymes (GST) have been shown to catalyze protein S-glutathionylation, and the glutaredoxin Grx1 is the main responsible for protein deglutathionylation in mammalian cells, with higher activity than the thioredoxin system [9]. Despite a growing interest of scientists in unraveling regulatory mechanisms of protein glutathionylation, the amount of all atom molecular dynamics studies present in the literature remains scarce compared to other mainstream PTM such as phosphorylation [18]. In most recent computational studies, Hyslop et al reported the structural impact of SSG on the GADPH protein structure [19], Moffet et al described the SSG-induced allosteric effects in a plant kinase [20], Zhou et al investigated the effect of SSG on ATP/ADP binding to the adenine nucleotide translocator 2 protein [21], Zaffagnini et al provided clues on SSG triggered collapse of a soluble tetrameric plant protein [22], and Hameed studied the inhibition of triose-phosphate isomerase by SSG [23].…”

Section: Introductionmentioning

confidence: 99%

Robust of AMBER force field parameters for glutathionylated cysteines

Elftmaoui,

Bignon

2023

Preprint

View full text Add to dashboard Cite

S-glutathionylation is an oxidative post-translational modification which is involved in the regulation of many cell signaling pathways. Increasing amounts of studies show that it is crucial in cell homeostasis and often deregulated in several pathologies. However, the effect of S-glutathionylation on proteins structure and activity is poorly understood, and a drastic lack of structural information at the atomic scale remains. Molecular dynamics simulations, which can provide important information about modification-induced modulation of proteins structure and function, are also sparse and there is no benchmarked force field parameters for this modified cysteine. In this contribution, we provide robust AMBER parameters for S-glutathionylation, that we tested extensively against experimental data through a total of 33 micro sec molecular dynamics simulations. We show that our parameters set efficiently describe the global and local structural properties of S-glutathionylated proteins. These data provide the community with an important tool to stimulate investigations about the effect of S-glutathionylation on protein dynamics and function, in a common effort to unravel the structural mechanisms underlying their critical role in cellular processes.

show abstract

Thirty years of molecular dynamics simulations on posttranslational modifications of proteins

Abstract: Posttranslational modifications (PTMs) are an integral component to how cells respond to perturbation. While experimental advances have enabled improved PTM identification capabilities, the same throughput for characterizing how structural changes...

Cited by 15 publications

References 285 publications

Active Learning of the Conformational Ensemble of Proteins Using Maximum Entropy VAMPNets

Active Learning of the Conformational Ensemble of Proteins Using Maximum Entropy VAMPNets

Active Learning of the Conformational Ensemble of Proteins using Maximum Entropy VAMPNets

Robust of AMBER force field parameters for glutathionylated cysteines

Contact Info

Product

Resources

About