Protein storytelling through physics

Advancing Molecular Dynamics: Toward Standardization, Integration, and Data Accessibility in Structural Biology

Molecular dynamics (MD) simulations have become a valuable tool in structural biology, offering insights into complex biological systems that are difficult to obtain through experimental techniques alone. The lack of available data sets and structures in most published computational work has limited other researchers' use of these models. In recent years, the emergence of online sharing platforms and MD database initiatives favor the deposition of ensembles and structures to accompany publications, favoring reuse of the data sets. However, the lack of uniform metadata collection, formats, and what data are deposited limits the impact and its use by different communities that are not necessarily experts in MD. This Perspective highlights the need for standardization and better resource sharing for processing and interpreting MD simulation results, akin to efforts in other areas of structural biology. As the field moves forward, we will see an increase in popularity and benefits of MD-based integrative approaches combining experimental data and simulations through probabilistic reasoning, but these too are limited by uniformity in experimental data availability and choices on how the data are modeled that are not trivial to decipher from papers. Other fields have addressed similar challenges comprehensively by establishing task forces with different degrees of success. The large scope and number of communities to represent the breadth of types of MD simulations complicates a parallel approach that would fit all. Thus, each group typically decides what data and which format to upload on servers like Zenodo. Uploading data with FAIR (findable, accessible, interoperable, reusable) principles in mind including optimal metadata collection will make the data more accessible and actionable by the community. Such a wealth of simulation data will foster method development and infrastructure advancements, thus propelling the field forward.

Accelerating Kinetics with Time-Reversal Path Sampling

In comparison to numerous enhanced sampling methods for equilibrium thermodynamics, accelerating simulations for kinetics and nonequilibrium statistics are relatively rare and less effective. Here, we derive a time-reversal path sampling (tRPS) method based on time reversibility to accelerate simulations for determining the transition rates between free-energy basins. It converts the difficult uphill path sampling into an easy downhill problem. This method is easy to implement, i.e., forward and backward shooting simulations with opposite initial velocities are conducted from random initial conformations within a transition-state region until they reach the basin minima, which are then assembled to give the distribution of transition paths efficiently. The effects of tRPS are demonstrated using a comparison with direct simulations of protein folding and unfolding, where tRPS is shown to give results consistent with direct simulations and increase the efficiency by up to five orders of magnitude. This approach is generally applicable to stochastic processes with microscopic reversibility, regardless of whether the variables are continuous or discrete.

Rapid Rescoring and Refinement of Ligand-Receptor Complexes Using Replica Exchange Molecular Dynamics with a Monte Carlo Pose Reservoir

Virtual screening (VS) involves generation of poses for a library of ligands and ranking using simplified energy functions and limited flexibility. Top-scored poses are used to rank and prioritize ligands. Here, we adapt the reservoir replica exchange molecular dynamics (res-REMD) method to rerank poses generated through VS. REMD simulations are carried out but with occasional Monte Carlo jumps to alternate VS-generated poses using a Metropolis criterion. The simulations converge within 10 ns for all systems, generating populations of alternate poses in the context of fully flexible ligand and protein side chains. The protocol is applied to four model protein-ligand complexes, where DOCK resulted in two successes and two scoring failures. In all four systems, the most populated cluster from the final ensemble exhibits high similarity to the crystallographic pose with ligand RMSD values under 2.0 Å. Both DOCK failures were rescued. For one DOCK success, the protocol identified the correct pose but also sampled an alternate pose at equal probability. Opportunities for future improvements and extensions are discussed.

Enabling technology and core theory of synthetic biology

Synthetic biology provides a new paradigm for life science research ("build to learn") and opens the future journey of biotechnology ("build to use"). Here, we discuss advances of various principles and technologies in the mainstream of the enabling technology of synthetic biology, including synthesis and assembly of a genome, DNA storage, gene editing, molecular evolution and de novo design of function proteins, cell and gene circuit engineering, cell-free synthetic biology, artificial intelligence (AI)-aided synthetic biology, as well as biofoundries. We also introduce the concept of quantitative synthetic biology, which is guiding synthetic biology towards increased accuracy and predictability or the real rational design. We conclude that synthetic biology will establish its disciplinary system with the iterative development of enabling technologies and the maturity of the core theory.

The nucleotide excision repair proteins through the lens of molecular dynamics simulations

Mutations that affect the proteins responsible for the nucleotide excision repair (NER) pathway can lead to diseases such as xeroderma pigmentosum, trichothiodystrophy, Cockayne syndrome, and Cerebro-oculo-facio-skeletal syndrome. Hence, understanding their molecular behavior is needed to elucidate these diseases' phenotypes and how the NER pathway is organized and coordinated. Molecular dynamics techniques enable the study of different protein conformations, adaptable to any research question, shedding light on the dynamics of biomolecules. However, as important as they are, molecular dynamics studies focused on DNA repair pathways are still becoming more widespread. Currently, there are no review articles compiling the advancements made in molecular dynamics approaches applied to NER and discussing: (i) how this technique is currently employed in the field of DNA repair, focusing on NER proteins; (ii) which technical setups are being employed, their strengths and limitations; (iii) which insights or information are they providing to understand the NER pathway or NER-associated proteins; (iv) which open questions would be suited for this technique to answer; and (v) where can we go from here. These questions become even more crucial considering the numerous 3D structures published regarding the NER pathway's proteins in recent years. In this work, we tackle each one of these questions, revising and critically discussing the results published in the context of the NER pathway.

MELD-Bracket Ranks Binding Affinities of Diverse Sets of Ligands

Affinity ranking of structurally diverse small-molecule ligands is a challenging problem with important applications in structure-based drug discovery. Absolute binding free energy methods can model diverse ligands, but the high computational cost of the current methods limits application to data sets with few ligands. We recently developed MELD-Bracket, a Molecular Dynamics method for efficient affinity ranking of ligands [ JCTC 2022, 18 (1), 374-379]. It utilizes a Bayesian framework to guide sampling to relevant regions of phase space, and it couples this with a bracket-like competition on a pool of ligands. Here we find that 6-competitor MELD-Bracket can rank dozens of diverse ligands that have low structural similarity and different net charges. We benchmark it on four protein systems─PTB1B, Tyk2, BACE, and JAK3─having varied modes of interactions. We also validated 8-competitor and 12-competitor protocols. The MELD-Bracket protocols presented here may have the appropriate balance of accuracy and computational efficiency to be suitable for ranking diverse ligands from typical drug discovery campaigns.

Exploring the Transferability of Replica Exchange Structure Reservoirs to Accelerate Generation of Ensembles for Alternate Hamiltonians or Protein Mutations

Generating precise ensembles is commonly a prerequisite to understand the energetics of biological processes using Molecular Dynamics (MD) simulations. Previously, we have shown how unweighted reservoirs built from high temperature MD simulations can accelerate convergence of Boltzmann-weighted ensembles by at least 10× with the Reservoir Replica Exchange MD (RREMD) method. Therefore, in this work, we explore whether an unweighted structure reservoir generated with one Hamiltonian (solute force field plus solvent model) can be reused to quickly generate accurately weighted ensembles for Hamiltonians other than the one that was used to generate the reservoir. We also extended this methodology to rapidly estimate the effects of mutations on peptide stability by using a reservoir of diverse structures obtained from wild-type simulations. These results suggest that structures generated via fast methods such as coarse-grained models or structures predicted by Rosetta or deep learning approaches could be integrated into a reservoir to accelerate generation of ensembles using more accurate representations.

Protein structure prediction using the evolutionary algorithm USPEX

Protein structure prediction is one of major problems of modern biophysics: current attempts to predict the tertiary protein structure from amino acid sequence are successful mostly when the use of big data and machine learning allows one to reduce the "prediction problem" to the "problem of recognition". Compared with recent successes of deep learning, classical predictive methods lag behind in their accuracy for the prediction of stable conformations. Therefore, in this work we extended the evolutionary algorithm USPEX to predict protein structure based on global optimization starting with the amino acid sequence. Moreover, we compared frequently used force fields for the task of protein structure prediction. Protein structure relaxation and energy calculations were performed using Tinker (with several different force fields) and Rosetta (with REF2015 force field) codes. To create new protein structure models in the USPEX algorithm, we developed novel variation operators. The test of the new method on seven proteins having (for simplicity) no cis-proline (with ω ≈ 0°) residues, and a length of up to 100 residues, revealed that our algorithm predicts tertiary structures of proteins with high accuracy. The comparison of the final potential energies of the predicted protein structures obtained using the USPEX and the Rosetta Abinitio approach showed that in most cases the developed algorithm found structures with close or even lower energy (Amber/Charmm/Oplsaal) and scoring function (REF2015). While USPEX has clearly demonstrated its ability to find very deep energy minima, our study showed that the existing force fields are not sufficiently accurate for accurate blind prediction of protein structures without further experimental verification.

Stability and conformational memory of electrosprayed and rehydrated bacteriophage MS2 virus coat proteins

Proteins are innately dynamic, which is important for their functions, but which also poses significant challenges when studying their structures. Gas-phase techniques can utilise separation and a range of sample manipulations to transcend some of the limitations of conventional techniques for structural biology in crystalline or solution phase, and isolate different states for separate interrogation. However, the transfer from solution to the gas phase risks affecting the structures, and it is unclear to what extent different conformations remain distinct in the gas phase, and if resolution in silico can recover the native conformations and their differences. Here, we use extensive molecular dynamics simulations to study the two distinct conformations of dimeric capsid protein of the MS2 bacteriophage. The protein undergoes notable restructuring of its peripheral parts in the gas phase, but subsequent simulation in solvent largely recovers the native structure. Our results suggest that despite some structural loss due to the experimental conditions, gas-phase structural biology techniques provide meaningful data that inform not only about the structures but also conformational dynamics of proteins.

•

Presented extensive molecular dynamics (MD) simulation data investigating protein vacuum exposure and rehydration dynamics.

•

Demonstrated that the majority of the protein structure recovers their initial solution conformation after vacuum exposure.

•

Explored the potential gain for structural biology of using MD simulation to refine gas-phase determined protein structures.

Mechanistic basis of the increased methylation activity of the SETD2 protein lysine methyltransferase towards a designed super-substrate peptide

Protein lysine methyltransferases have important regulatory functions in cells, but mechanisms determining their activity and specificity are incompletely understood. Naturally, SETD2 introduces H3K36me3, but previously an artificial super-substrate (ssK36) was identified, which is methylated >100-fold faster. The ssK36-SETD2 complex structure cannot fully explain this effect. We applied molecular dynamics (MD) simulations and biochemical experiments to unravel the mechanistic basis of the increased methylation of ssK36, considering peptide conformations in solution, association of peptide and enzyme, and formation of transition-state (TS) like conformations of the enzyme-peptide complex. We observed in MD and FRET experiments that ssK36 adopts a hairpin conformation in solution with V35 and K36 placed in the loop. The hairpin conformation has easier access into the active site of SETD2 and it unfolds during the association process. Peptide methylation experiments revealed that introducing a stable hairpin conformation in the H3K36 peptide increased its methylation by SETD2. In MD simulations of enzyme-peptide complexes, the ssK36 peptide approached TS-like structures more frequently than H3K36 and distinct, substrate-specific TS-like structures were observed. Hairpin association, hairpin unfolding during association, and substrate-specific catalytically competent conformations may also be relevant for other PKMTs and hairpins could represent a promising starting point for SETD2 inhibitor development.

Dissecting the stability determinants of a challenging de novo protein fold using massively parallel design and experimentation

Designing entirely new protein structures remains challenging because we do not fully understand the biophysical determinants of folding stability. Yet, some protein folds are easier to design than others. Previous work identified the 43-residue ɑββɑ fold as especially challenging: The best designs had only a 2% success rate, compared to 39 to 87% success for other simple folds [G. J. Rocklin et al., Science 357, 168-175 (2017)]. This suggested the ɑββɑ fold would be a useful model system for gaining a deeper understanding of folding stability determinants and for testing new protein design methods. Here, we designed over 10,000 new ɑββɑ proteins and found over 3,000 of them to fold into stable structures using a high-throughput protease-based assay. NMR, hydrogen-deuterium exchange, circular dichroism, deep mutational scanning, and scrambled sequence control experiments indicated that our stable designs fold into their designed ɑββɑ structures with exceptional stability for their small size. Our large dataset enabled us to quantify the influence of universal stability determinants including nonpolar burial, helix capping, and buried unsatisfied polar atoms, as well as stability determinants unique to the ɑββɑ topology. Our work demonstrates how large-scale design and test cycles can solve challenging design problems while illuminating the biophysical determinants of folding.

SP-A binding to the SARS-CoV-2 spike protein using hybrid quantum and classical in silico modeling and molecular pruning by Quantum Approximate Optimization Algorithm (QAOA) Based MaxCut with ZDOCK

The pulmonary surfactant protein A (SP-A) is a constitutively expressed immune-protective collagenous lectin (collectin) in the lung. It binds to the cell membrane of immune cells and opsonizes infectious agents such as bacteria, fungi, and viruses through glycoprotein binding. SARS-CoV-2 enters airway epithelial cells by ligating the Angiotensin Converting Enzyme 2 (ACE2) receptor on the cell surface using its Spike glycoprotein (S protein). We hypothesized that SP-A binds to the SARS-CoV-2 S protein and this binding interferes with ACE2 ligation. To study this hypothesis, we used a hybrid quantum and classical in silico modeling technique that utilized protein graph pruning. This graph pruning technique determines the best binding sites between amino acid chains by utilizing the Quantum Approximate Optimization Algorithm (QAOA)-based MaxCut (QAOA-MaxCut) program on a Near Intermediate Scale Quantum (NISQ) device. In this, the angles between every neighboring three atoms were Fourier-transformed into microwave frequencies and sent to a quantum chip that identified the chemically irrelevant atoms to eliminate based on their chemical topology. We confirmed that the remaining residues contained all the potential binding sites in the molecules by the Universal Protein Resource (UniProt) database. QAOA-MaxCut was compared with GROMACS with T-REMD using AMBER, OPLS, and CHARMM force fields to determine the differences in preparing a protein structure docking, as well as with Goemans-Williamson, the best classical algorithm for MaxCut. The relative binding affinity of potential interactions between the pruned protein chain residues of SP-A and SARS-CoV-2 S proteins was assessed by the ZDOCK program. Our data indicate that SP-A could ligate the S protein with a similar affinity to the ACE2-Spike binding. Interestingly, however, the results suggest that the most tightly-bound SP-A binding site is localized to the S2 chain, in the fusion region of the SARS-CoV-2 S protein, that is responsible for cell entry Based on these findings we speculate that SP-A may not directly compete with ACE2 for the binding site on the S protein, but interferes with viral entry to the cell by hindering necessary conformational changes or the fusion process.

Modified Protein-Water Interactions in CHARMM36m for Thermodynamics and Kinetics of Proteins in Dilute and Crowded Solutions

Proper balance between protein-protein and protein-water interactions is vital for atomistic molecular dynamics (MD) simulations of globular proteins as well as intrinsically disordered proteins (IDPs). The overestimation of protein-protein interactions tends to make IDPs more compact than those in experiments. Likewise, multiple proteins in crowded solutions are aggregated with each other too strongly. To optimize the balance, Lennard-Jones (LJ) interactions between protein and water are often increased about 10% (with a scaling parameter, λ = 1.1) from the existing force fields. Here, we explore the optimal scaling parameter of protein-water LJ interactions for CHARMM36m in conjunction with the modified TIP3P water model, by performing enhanced sampling MD simulations of several peptides in dilute solutions and conventional MD simulations of globular proteins in dilute and crowded solutions. In our simulations, 10% increase of protein-water LJ interaction for the CHARMM36m cannot maintain stability of a small helical peptide, (AAQAA)₃ in a dilute solution and only a small modification of protein-water LJ interaction up to the 3% increase (λ = 1.03) is allowed. The modified protein-water interactions are applicable to other peptides and globular proteins in dilute solutions without changing thermodynamic properties from the original CHARMM36m. However, it has a great impact on the diffusive properties of proteins in crowded solutions, avoiding the formation of too sticky protein-protein interactions.

Accelerating the Ensemble Convergence of RNA Hairpin Simulations with a Replica Exchange Structure Reservoir

RNA is a key participant in many biological processes, but studies of RNA using computer simulations lag behind those of proteins, largely due to less-developed force fields and the slow dynamics of RNA. Generating converged RNA ensembles for force field development and other studies remains a challenge. In this study, we explore the ability of replica exchange molecular dynamics to obtain well-converged conformational ensembles for two RNA hairpin systems in an implicit solvent. Even for these small model systems, standard REMD remains computationally costly, but coupling to a pre-generated structure library using the reservoir REMD approach provides a dramatic acceleration of ensemble convergence for both model systems. Such precise ensembles could facilitate RNA force field development and validation and applications of simulation to more complex RNA systems. The advantages and remaining challenges of applying R-REMD to RNA are investigated in detail.

Computational Studies of Lipid Droplets

Lipid droplets (LDs) are intracellular organelles whose primary function is energy storage. Known to emerge from the endoplasmic reticulum (ER) bilayer, LDs have a unique structure with a core consisting of neutral lipids, triacylglycerol (TG) or sterol esters (SE), surrounded by a phospholipid (PL) monolayer and decorated by proteins that come and go throughout their complex lifecycle. In this Feature Article, we review recent developments in computational studies of LDs, a rapidly growing area of research. We highlight how molecular dynamics (MD) simulations have provided valuable molecular-level insight into LD targeting and LD biogenesis. Additionally, we review the physical properties of TG from different force fields compared with experimental data. Possible future directions and challenges are discussed.

Heterogeneity in Protein Folding and Unfolding Reactions

Proteins have dynamic structures that undergo chain motions on time scales spanning from picoseconds to seconds. Resolving the resultant conformational heterogeneity is essential for gaining accurate insight into fundamental mechanistic aspects of the protein folding reaction. The use of high-resolution structural probes, sensitive to population distributions, has begun to enable the resolution of site-specific conformational heterogeneity at different stages of the folding reaction. Different states populated during protein folding, including the unfolded state, collapsed intermediate states, and even the native state, are found to possess significant conformational heterogeneity. Heterogeneity in protein folding and unfolding reactions originates from the reduced cooperativity of various kinds of physicochemical interactions between various structural elements of a protein, and between a protein and solvent. Heterogeneity may arise because of functional or evolutionary constraints. Conformational substates within the unfolded state and the collapsed intermediates that exchange at rates slower than the subsequent folding steps give rise to heterogeneity on the protein folding pathways. Multiple folding pathways are likely to represent distinct sequences of structure formation. Insight into the nature of the energy barriers separating different conformational states populated during (un)folding can also be obtained by resolving heterogeneity.

Accelerating Protein Folding Molecular Dynamics Using Inter-Residue Distances from Machine Learning Servers

Recently, predicting the native structures of proteins has become possible using computational molecular physics (CMP)─physics-based force fields sampled with proper statistics─but only for small proteins. Algorithms with better scaling are needed. We describe ML x MELD x MD, a molecular dynamics (MD) method that inputs residue contacts derived from machine learning (ML) servers into MELD, a Bayesian accelerator that preserves detailed-balance statistics. Contacts are derived from trRosetta-predicted distance histograms (distograms) and are integrated into MELD's atomistic MD as spatial restraints through parametrized potential functions. In the CASP14 blind prediction event, ML x MELD x MD predicted 13 native structures to better than 4.5 Å error, including for 10 proteins in the range of 115-250 amino acids long. Also, the scaling of simulation time vs protein length is much better than unguided MD: tsim ∼ e0.023N for ML x MELD x MD vs tsim ∼ e0.168N for MD alone. This shows how machine learning information can be leveraged to advance physics-based modeling of proteins.

An empirical energy landscape reveals mechanism of proteasome in polypeptide translocation

The ring-like ATPase complexes in the AAA+ family perform diverse cellular functions that require coordination between the conformational transitions of their individual ATPase subunits (Erzberger and Berger, 2006; Puchades et al., 2020). How the energy from ATP hydrolysis is captured to perform mechanical work by these coordinated movements is unknown. In this study, we developed a novel approach for delineating the nucleotide-dependent free-energy landscape (FEL) of the proteasome’s heterohexameric ATPase complex based on complementary structural and kinetic measurements. We used the FEL to simulate the dynamics of the proteasome and quantitatively evaluated the predicted structural and kinetic properties. The FEL model predictions are consistent with a wide range of experimental observations in this and previous studies and suggested novel mechanistic features of the proteasomal ATPases. We find that the cooperative movements of the ATPase subunits result from the design of the ATPase hexamer entailing a unique free-energy minimum for each nucleotide-binding status. ATP hydrolysis dictates the direction of substrate translocation by triggering an energy-dissipating conformational transition of the ATPase complex.

In cells, many biological processes are carried out by large complexes made up of different proteins. These macromolecules act like miniature machines, flexing and moving their various parts to perform their cellular roles. One such complex is the 26S proteasome, which is responsible for recycling other proteins in the cell. The proteasome consists of approximately 31 subunits, including a ring of six ATPase enzymes that provide the complex with the energy it needs to mechanically unfold proteins.

To understand how the proteasome and other large complexes work, researchers need to be able to monitor how their structure changes over time. These dynamics are challenging to probe directly with experiments, but can be assessed using computer simulations which track the movement of individual molecules and atoms. However, currently available computer systems do not have enough power to simulate the dynamics of large protein assemblies, like the 26S proteasome: for example, it would take longer than a thousand years to model how each atom in the complex moves over a timescale in which a biological change would happen (roughly 100ms).

Here, Fang, Hon et al. have developed a new approach to simulate the structural dynamics of the proteasome’s ring of ATPase enzymes. Different known structures of the proteasome were used to identify the range of possible movements and shapes the complex can make. Fang, Hon et al. then used this data to calculate the energy level of each structure – also known as the ‘free energy landscape’ – and the rate of transition between them. This made it possible to simulate how the different ATPase enzymes move within the ring under a wide range of conditions.

The simulated ATPase movements predicted how the proteasome machine would behave during various tasks, including degrading other proteins. Fan, Hon et al. carefully examined these predictions and found that they were consistent with experimental observations, validating their new simulation method.

This work demonstrates the feasibility of simulating the actions of a large protein complex based on its free energy landscape. The results offer important insights into the functional mechanics of the 26S proteasome and related protein machines. Further work may help to simplify this process so the approach can be used to investigate the dynamics of other protein assemblies.

Accelerating Molecular Dynamics Enrichments of High-Affinity Ligands for Proteins

Molecular docking algorithms are used to seek the most active compounds from a pool of ligands. In principle, molecular dynamics (MD) simulations with accurate physical potentials and sampling could yield better enrichments, but they are computationally expensive. Here, we describe a method called MELD-Bracket that utilizes biased replica exchange ladders in MD in order to compete different ligands against each other within a fast bracket style "binding tournament". MELD-Bracket finds best-binders rapidly when ligands are well separated in their binding affinities.

From crystallographic data to the solution structure of photoreceptors: the case of the AppA BLUF domain

Photoreceptor proteins bind a chromophore, which, upon light absorption, modifies its geometry or its interactions with the protein, finally inducing the structural change needed to switch the protein from an inactive to an active or signaling state. In the Blue Light-Using Flavin (BLUF) family of photoreceptors, the chromophore is a flavin and the changes have been connected with a rearrangement of the hydrogen bond network around it on the basis of spectroscopic changes measured for the dark-to-light conversion. However, the exact conformational change triggered by the photoexcitation is still elusive mainly because a clear consensus on the identity not only of the light activated state but also of the dark one has not been achieved. Here, we present an integrated investigation that combines microsecond MD simulations starting from the two conflicting crystal structures available for the AppA BLUF domain with calculations of NMR, IR and UV-Vis spectra using a polarizable QM/MM approach. Thanks to such a combined analysis of the three different spectroscopic responses, a robust characterization of the structure of the dark state in solution is given together with the uncovering of important flaws of the most popular molecular mechanisms present in the literature for the dark-to-light activation.

Using a Coarse-Grained Modeling Framework to Identify Oligomeric Motifs with Tunable Secondary Structure

Coarse-grained modeling can be used to explore general theories that are independent of specific chemical detail. In this paper, we present cg_openmm, a Python-based simulation framework for modeling coarse-grained hetero-oligomers and screening them for structural and thermodynamic characteristics of cooperative secondary structures. cg_openmm facilitates the building of coarse-grained topology and random starting configurations, setup of GPU-accelerated replica exchange molecular dynamics simulations with the OpenMM software package, and features a suite of postprocessing thermodynamic and structural analysis tools. In particular, native contact analysis, heat capacity calculations, and free energy of folding calculations are used to identify and characterize cooperative folding transitions and stable secondary structures. In this work, we demonstrate the capabilities of cg_openmm on a simple 1-1 Lennard-Jones coarse-grained model, in which each residue contains 1 backbone and 1 side-chain bead. By scanning both nonbonded and bonded force-field parameter spaces at the coarse-grained level, we identify and characterize sets of parameters which result in the formation of stable helices through cooperative folding transitions. Moreover, we show that the geometries and stabilities of these helices can be tuned by manipulating the force-field parameters.

The Protein Folding Problem: The Role of Theory

The protein folding problem was first articulated as question of how order arose from disorder in proteins: How did the various native structures of proteins arise from interatomic driving forces encoded within their amino acid sequences, and how did they fold so fast? These matters have now been largely resolved by theory and statistical mechanics combined with experiments. There are general principles. Chain randomness is overcome by solvation-based codes. And in the needle-in-a-haystack metaphor, native states are found efficiently because protein haystacks (conformational ensembles) are funnel-shaped. Order-disorder theory has now grown to encompass a large swath of protein physical science across biology.

Contemporary biomedical engineering perspective on volitional evolution for human radiotolerance enhancement beyond low-earth orbit

A primary objective of the National Aeronautics and Space Administration (NASA) is expansion of humankind's presence outside low-Earth orbit, culminating in permanent interplanetary travel and habitation. Having no inherent means of physiological detection or protection against ionizing radiation, humans incur capricious risk when journeying beyond low-Earth orbit for long periods. NASA has made large investments to analyze pathologies from space radiation exposure, emphasizing the importance of characterizing radiation's physiological effects. Because natural evolution would require many generations to confer resistance against space radiation, immediately pragmatic approaches should be considered. Volitional evolution, defined as humans steering their own heredity, may inevitably retrofit the genome to mitigate resultant pathologies from space radiation exposure. Recently, uniquely radioprotective genes have been identified, conferring local or systemic radiotolerance when overexpressed in vitro and in vivo. Aiding in this process, the CRISPR/Cas9 technique is an inexpensive and reproducible instrument capable of making limited additions and deletions to the genome. Although cohorts can be identified and engineered to protect against radiation, alternative and supplemental strategies should be seriously considered. Advanced propulsion and mild synthetic torpor are perhaps the most likely to be integrated. Interfacing artificial intelligence with genetic engineering using predefined boundary conditions may enable the computational modeling of otherwise overly complex biological networks. The ethical context and boundaries of introducing genetically pioneered humans are considered.

Differentiable molecular simulation can learn all the parameters in a coarse-grained force field for proteins

Finding optimal parameters for force fields used in molecular simulation is a challenging and time-consuming task, partly due to the difficulty of tuning multiple parameters at once. Automatic differentiation presents a general solution: run a simulation, obtain gradients of a loss function with respect to all the parameters, and use these to improve the force field. This approach takes advantage of the deep learning revolution whilst retaining the interpretability and efficiency of existing force fields. We demonstrate that this is possible by parameterising a simple coarse-grained force field for proteins, based on training simulations of up to 2,000 steps learning to keep the native structure stable. The learned potential matches chemical knowledge and PDB data, can fold and reproduce the dynamics of small proteins, and shows ability in protein design and model scoring applications. Problems in applying differentiable molecular simulation to all-atom models of proteins are discussed along with possible solutions and the variety of available loss functions. The learned potential, simulation scripts and training code are made available at https://github.com/psipred/cgdms.

The Proteome Folding Problem and Cellular Proteostasis

Stunning advances have been achieved in addressing the protein folding problem, providing deeper understanding of the mechanisms by which proteins navigate energy landscapes to reach their native states and enabling powerful algorithms to connect sequence to structure. However, the realities of the in vivo protein folding problem remain a challenge to reckon with. Here, we discuss the concept of the "proteome folding problem"-the problem of how organisms build and maintain a functional proteome-by admitting that folding energy landscapes are characterized by many misfolded states and that cells must deploy a network of chaperones and degradation enzymes to minimize deleterious impacts of these off-pathway species. The resulting proteostasis network is an inextricable part of in vivo protein folding and must be understood in detail if we are to solve the proteome folding problem. We discuss how the development of computational models for the proteostasis network's actions and the relationship to the biophysical properties of the proteome has begun to offer new insights and capabilities.

Highly accurate protein structure prediction with AlphaFold

Proteins are essential to life, and understanding their structure can facilitate a mechanistic understanding of their function. Through an enormous experimental effort1-4, the structures of around 100,000 unique proteins have been determined5, but this represents a small fraction of the billions of known protein sequences6,7. Structural coverage is bottlenecked by the months to years of painstaking effort required to determine a single protein structure. Accurate computational approaches are needed to address this gap and to enable large-scale structural bioinformatics. Predicting the three-dimensional structure that a protein will adopt based solely on its amino acid sequence-the structure prediction component of the 'protein folding problem'8-has been an important open research problem for more than 50 years9. Despite recent progress10-14, existing methods fall far short of atomic accuracy, especially when no homologous structure is available. Here we provide the first computational method that can regularly predict protein structures with atomic accuracy even in cases in which no similar structure is known. We validated an entirely redesigned version of our neural network-based model, AlphaFold, in the challenging 14th Critical Assessment of protein Structure Prediction (CASP14)15, demonstrating accuracy competitive with experimental structures in a majority of cases and greatly outperforming other methods. Underpinning the latest version of AlphaFold is a novel machine learning approach that incorporates physical and biological knowledge about protein structure, leveraging multi-sequence alignments, into the design of the deep learning algorithm.

Protein structure prediction by AlphaFold2: are attention and symmetries all you need?

The functions of most proteins result from their 3D structures, but determining their structures experimentally remains a challenge, despite steady advances in crystallography, NMR and single-particle cryoEM. Computationally predicting the structure of a protein from its primary sequence has long been a grand challenge in bioinformatics, intimately connected with understanding protein chemistry and dynamics. Recent advances in deep learning, combined with the availability of genomic data for inferring co-evolutionary patterns, provide a new approach to protein structure prediction that is complementary to longstanding physics-based approaches. The outstanding performance of AlphaFold2 in the recent Critical Assessment of protein Structure Prediction (CASP14) experiment demonstrates the remarkable power of deep learning in structure prediction. In this perspective, we focus on the key features of AlphaFold2, including its use of (i) attention mechanisms and Transformers to capture long-range dependencies, (ii) symmetry principles to facilitate reasoning over protein structures in three dimensions and (iii) end-to-end differentiability as a unifying framework for learning from protein data. The rules of protein folding are ultimately encoded in the physical principles that underpin it; to conclude, the implications of having a powerful computational model for structure prediction that does not explicitly rely on those principles are discussed.

A synergy between the catalytic and structural Zn(II) ions and the enzyme and substrate dynamics underlies the structure-function relationships of matrix metalloproteinase collagenolysis

Matrix metalloproteinases (MMPs) are Zn(II) dependent endopeptidases involved in the degradation of collagen. Unbalanced collagen breakdown results in numerous pathological conditions, including cardiovascular and neurodegenerative diseases and tumor growth and invasion. Matrix metalloproteinase-1 (MMP-1) is a member of the MMPs family. The enzyme contains catalytic and structural Zn(II) ions. Despite many studies on the enzyme, there is little known about the synergy between the two Zn(II) metal ions and the enzyme and substrate dynamics in MMP-1 structure-function relationships. We performed a computational study of the MMP-1•triple-helical peptide (THP) enzyme•substrate complex to provide this missing insight. Our results revealed Zn(II) ions' importance in modulating the long-range correlated motions in the MMP-1•THP complex. Overall, our results reveal the importance of the catalytic Zn(II) and the role of the structural Zn(II) ion in preserving the integrity of the enzyme active site and the overall enzyme-substrate complex synergy with the dynamics of the enzyme and the substrate. Notably, both Zn(II) sites participate in diverse networks of long-range correlated motions that involve the CAT and HPX domains and the THP substrate, thus exercising a complex role in the stability and functionality of the MMP-1•THP complex. Both the Zn(II) ions have a distinct impact on the structural stability and dynamics of the MMP-1•THP complex. The study shifts the paradigm from the "local role" of the Zn(II) ions with knowledge about their essential role in the long-range dynamics and stability of the overall enzyme•substrate (ES) complex.

Physical Characterization of Triolein and Implications for Its Role in Lipid Droplet Biogenesis

Lipid droplets (LDs) are neutral lipid-storing organelles surrounded by a phospholipid (PL) monolayer. At present, how LDs are formed in the endoplasmic reticulum (ER) bilayer is poorly understood. In this study, we present a revised all-atom (AA) triolein (TG) model, the main constituent of the LD core, and characterize its properties in a bilayer membrane to demonstrate the implications of its behavior in LD biogenesis. In bilayer simulations, TG resides at the surface, adopting PL-like conformations (denoted in this work as SURF-TG). Free energy sampling simulation results estimate the barrier for TG relocating from the bilayer surface to the bilayer center to be ∼2 kcal/mol in the absence of an oil lens. SURF-TG is able to modulate membrane properties by increasing PL ordering, decreasing bending modulus, and creating local negative curvature. The other neutral lipid, dioleoyl-glycerol (DAG), also reduces the membrane bending modulus and populates negative curvature regions. A phenomenological coarse-grained (CG) model is also developed to observe larger-scale SURF-TG-mediated membrane deformation. CG simulations confirm that TG nucleates between the bilayer leaflets at a critical concentration when SURF-TG is evenly distributed. However, when one monolayer contains more SURF-TG, the membrane bends toward the other leaflet, followed by TG nucleation if a concentration is higher than the critical threshold. The central conclusion of this study is that SURF-TG is a negative curvature inducer, as well as a membrane modulator. To this end, a model is proposed in which the accumulation of SURF-TG in the luminal leaflet bends the ER bilayer toward the cytosolic side, followed by TG nucleation.

Exploring the Minimum-Energy Pathways and Free-Energy Profiles of Enzymatic Reactions with QM/MM Calculations

Understanding molecular mechanisms of enzymatic reactions is of vital importance in biochemistry and biophysics. Here, we introduce new functions of hybrid quantum mechanical/molecular mechanical (QM/MM) calculations in the GENESIS program to compute the minimum-energy pathways (MEPs) and free-energy profiles of enzymatic reactions. For this purpose, an interface in GENESIS is developed to utilize a highly parallel electronic structure program, QSimulate-QM (https://qsimulate.com), calling it as a shared library from GENESIS. Second, algorithms to search the MEP are implemented, combining the string method (E et al. J. Chem. Phys. 2007, 126, 164103) with the energy minimization of the buffer MM region. The method implemented in GENESIS is applied to an enzyme, triosephosphate isomerase, which converts dihyroxyacetone phosphate to glyceraldehyde 3-phosphate in four proton-transfer processes. QM/MM-molecular dynamics simulations show performances of greater than 1 ns/day with the density functional tight binding (DFTB), and 10-30 ps/day with the hybrid density functional theory, B3LYP-D3. These performances allow us to compute not only MEP but also the potential of mean force (PMF) of the enzymatic reactions using the QM/MM calculations. The barrier height obtained as 13 kcal mol-1 with B3LYP-D3 in the QM/MM calculation is in agreement with the experimental results. The impact of conformational sampling in PMF calculations and the level of electronic structure calculations (DFTB vs B3LYP-D3) suggests reliable computational protocols for enzymatic reactions without high computational costs.

Fighting viruses with computers, right now

The synergistic conjunction of various technological revolutions with the accumulated knowledge and workflows is rapidly transforming several scientific fields. Particularly, Virology can now feed from accurate physical models, polished computational tools, and massive computational power to readily integrate high-resolution structures into biological representations of unprecedented detail. That preparedness allows for the first time to get crucial information for vaccine and drug design from in-silico experiments against emerging pathogens of worldwide concern at relevant action windows. The present work reviews some of the main milestones leading to these breakthroughs in Computational Virology, providing an outlook for future developments in capacity building and accessibility to computational resources.

Biomolecular Modeling and Simulation: A Prospering Multidisciplinary Field

We reassess progress in the field of biomolecular modeling and simulation, following up on our perspective published in 2011. By reviewing metrics for the field's productivity and providing examples of success, we underscore the productive phase of the field, whose short-term expectations were overestimated and long-term effects underestimated. Such successes include prediction of structures and mechanisms; generation of new insights into biomolecular activity; and thriving collaborations between modeling and experimentation, including experiments driven by modeling. We also discuss the impact of field exercises and web games on the field's progress. Overall, we note tremendous success by the biomolecular modeling community in utilization of computer power; improvement in force fields; and development and application of new algorithms, notably machine learning and artificial intelligence. The combined advances are enhancing the accuracy andscope of modeling and simulation, establishing an exemplary discipline where experiment and theory or simulations are full partners.

Biomolecular modeling thrives in the age of technology

The biomolecular modeling field has flourished since its early days in the 1970s due to the rapid adaptation and tailoring of state-of-the-art technology. The resulting dramatic increase in size and timespan of biomolecular simulations has outpaced Moore's law. Here, we discuss the role of knowledge-based versus physics-based methods and hardware versus software advances in propelling the field forward. This rapid adaptation and outreach suggests a bright future for modeling, where theory, experimentation and simulation define three pillars needed to address future scientific and biomedical challenges.

Amyloid Oligomers: A Joint Experimental/Computational Perspective on Alzheimer's Disease, Parkinson's Disease, Type II Diabetes, and Amyotrophic Lateral Sclerosis

Protein misfolding and aggregation is observed in many amyloidogenic diseases affecting either the central nervous system or a variety of peripheral tissues. Structural and dynamic characterization of all species along the pathways from monomers to fibrils is challenging by experimental and computational means because they involve intrinsically disordered proteins in most diseases. Yet understanding how amyloid species become toxic is the challenge in developing a treatment for these diseases. Here we review what computer, in vitro, in vivo, and pharmacological experiments tell us about the accumulation and deposition of the oligomers of the (Aβ, tau), α-synuclein, IAPP, and superoxide dismutase 1 proteins, which have been the mainstream concept underlying Alzheimer's disease (AD), Parkinson's disease (PD), type II diabetes (T2D), and amyotrophic lateral sclerosis (ALS) research, respectively, for many years.

Protein Unfolding and Aggregation near a Hydrophobic Interface

The behavior of proteins near interfaces is relevant for biological and medical purposes. Previous results in bulk show that, when the protein concentration increases, the proteins unfold and, at higher concentrations, aggregate. Here, we study how the presence of a hydrophobic surface affects this course of events. To this goal, we use a coarse-grained model of proteins and study by simulations their folding and aggregation near an ideal hydrophobic surface in an aqueous environment by changing parameters such as temperature and hydrophobic strength, related, e.g., to ions concentration. We show that the hydrophobic surface, as well as the other parameters, affect both the protein unfolding and aggregation. We discuss the interpretation of these results and define future lines for further analysis, with their possible implications in neurodegenerative diseases.