Efficient Structure Prediction Research Articles

Frustraevo: a web server to localize and quantify the conservation of local energetic frustration in protein families.

According to the Principle of Minimal Frustration, folded proteins canonly have a minimal number of strong energetic conflicts in their native states. However, not all interactions are energetically optimized for folding but some remain in energetic conflict, i.e. they are highly frustrated. This remaining local energetic frustration has been shown to be statistically correlated with distinct functional aspects such as protein-protein interaction sites, allosterism and catalysis. Fuelled by the recent breakthroughs in efficient protein structure prediction that have made available good quality models for most proteins, we have developed a strategy to calculate local energetic frustration within large protein families and quantify its conservation over evolutionary time. Based on this evolutionary information we can identify how stability and functional constraints have appeared at the common ancestor of the family and have been maintained over the course of evolution. Here, we present FrustraEvo, a web server tool to calculate and quantify the conservation of local energetic frustration in protein families.

Superconductivity in (Sr, Ba)B3Si3 clathrate compounds

Extensive research has been conducted to explore sodalite clathrate compounds of groups 3 and 4, with various metal guest atoms, following the successful synthesis of XB3C3 (X = Sr, La) compounds. While the requirement for synthesis is still challenging. Here, using an efficient structure prediction method coupled with density functional theory, we investigated the stable phases and properties of MB3Si3 (M = Mg, Ca, Sr, and Ba) at both ambient and high pressure. Remarkably, our predictions uncover the presence of Pm-3n-SrB3Si3 and Pm-3n-BaB3Si3 compounds with clathrate-based sodalite structures, wherein the framework is composed of covalent B-Si bonds. Notably, compared to the elemental phases, BaB3Si3 has a negative enthalpy of formation at 1 atm, indicating the synthesis might be realized under conditions of ambient pressure. Additionally, according to calculations using electron–phonon coupling, the Pm-3n phase of SrB3Si3 and BaB3Si3 both have critical temperatures of 6.0 K and 3.9 K at atmospheric pressure, respectively, which show superconducting properties. Our research contributes to the understanding of Type-VII clathrate structure of alkaline-earth metal borosilicide that exhibit superconductivity.

Towards efficient structure prediction and pre-compensation in multi-photon lithography: publisher's note.

This publisher's note contains a correction to [Opt. Express30, 28805 (2022)10.1364/OE.462775].

Towards efficient structure prediction andpre-compensation in multi-photon lithography: publisher’s note

Towards efficient structure prediction andpre-compensation in multi-photon lithography: publisher’s note

Efficient Crystal Structure Prediction for Structurally Related Molecules with Accurate and Transferable Tailor-Made Force Fields.

Crystal structure prediction (CSP) is generally used to complement experimental solid form screening and applied to individual molecules in drug development. The fast development of algorithms and computing resources offers the opportunity to use CSP earlier and for a broader range of applications in the drug design cycle. This study presents a novel paradigm of CSP specifically designed for structurally related molecules, referred to as Quick-CSP. The approach prioritizes more accurate physics through robust and transferable tailor-made force fields (TMFFs), such that significant efficiency gains are achieved through the reduction of expensive ab initio calculations. The accuracy of the TMFF is increased by the introduction of electrostatic multipoles, and the fragment-based force field parameterization scheme is demonstrated to be transferable for a family of chemically related molecules. The protocol is benchmarked with structurally related compounds from the Bromodomain and Extraterminal (BET) domain inhibitors series. A new convergence criterion is introduced that aims at performing only as many ab initio optimizations of crystal structures as required to locate the bottom of the crystal energy landscape within a user-defined accuracy. The overall approach provides significant cost savings ranging from three- to eight-fold less than the full-CSP workflow. The reported advancements expand the scope and utility of the underlying CSP building blocks as well as their novel reassembly to other applications earlier in the drug design cycle to guide molecule design and selection.

Towards efficient structure prediction and pre-compensation in multi-photon lithography.

Microscale 3D printing technologies have been of increasing interest in industry and research for several years. Unfortunately, the fabricated structures always deviate from the respective expectations, often caused by the physico-chemical properties during and after the printing process. Here, we show first steps towards a simple, fast and easy to implement algorithm to predict the final structure topography for multi-photon lithography - also known as Direct Laser Writing (DLW). The three main steps of DLW, (i) exposure of a photo resin, (ii) cross-linking of the resin, and (iii) subsequent shrinkage are approximated by mathematical operations, showing promising results in coincidence with experimental observations. For example, the root-mean-square error (rmse) between the unmodified 3D print of a radial-symmetrically chirped topography and our predicted topography is only 0.46 µm, whereas the rmse between this 3D print and its target is 1.49 µm. Thus, our robust predictions can be used prior to the printing process to minimize undesired deviations between the target structure and the final 3D printed structure. Using a Downhill-Simplex algorithm for identifying the optimal prediction parameters, we were able to reduce the rmse from 4.04 µm to 0.33 µm by only two correction loops in our best-case scenario (rmse = 0.72 µm after one loop). Consequently, this approach can eliminate the need for many structural optimization loops to produce highly conformal and high quality micro structures in the future.

Toward the Rational Design of Mid‐Infrared Nonlinear Optical Materials with Targeted Properties via a Multi‐Level Data‐Driven Approach

AbstractDesign and exploratory synthesis of new mid‐infrared (mid‐IR) nonlinear optical (NLO) materials are urgently needed for modern laser science and technology because the widely used IR NLO crystals still suffer from their inextricable drawbacks. Herein, a multi‐level data‐driven approach to realize fast and efficient structure prediction for the exploration of promising mid‐IR NLO materials is proposed. Techniques based on machine learning, crystal structure prediction, high‐throughput calculation and screening, database building, and experimental verification are tightly combined for creating pathways from chemical compositions, crystal structures to rational synthesis. Through this data‐driven approach, not only are all known structures successfully predicted but also five thermodynamically stable and 50 metastable new selenides in AIBIIISe2 systems (AI = Li, Na, K, Rb, and Cs; BIII = Al and Ga) are found, among which eight outstanding compounds with wide bandgaps (> 2.70 eV) and large SHG responses (>10 pm V−1) are suggested. Moreover, the predicted compounds I2d‐LiGaSe2 and I4/mcm‐KAlSe2 are successfully obtained experimentally. In particular, LiGaSe2 exhibits a robust SHG response (≈2 × AGS) and long IR absorption edge that can cover two atmospheric windows (3–5, 8–12 µm). Simultaneously, this new research paradigm is also applicative for discovering new materials in other fields.

Physicochemical properties and homology studies of the floral meristem identity gene LFY in nonflowering and flowering plants.

In flowering plants, the LEAFY (LFY) gene controls floral meristem activity. In early land plants such as mosses and ferns, it, however, has a minimum role in cell division and development of diploid sporophyte. Homology modeling, an accurate and efficient protein structure prediction method, was used to construct a 3D model of the LEAFY protein in nonflowering and flowering plants. The present study examines the following species: Charophyte green algae, Physcomitrella, Ceratopteris, Picea, and Arabidopsis, as they are the popularly used model organisms for developmental studies. LEAFY protein sequences from the model organisms were aligned by multiple sequence alignment. 3D models of the LEAFY protein from all the model organisms was constructed using the PHYRE2 program with 100% confidence, and the constructed models were evaluated using the MolProbity tool. On the basis of the conserved regions, Charophyte green algae shared 38-46% sequence similarity with Physcomitrella sp., 37-46% similarity with Ceratopteris sp., 33-41% similarity with Picea sp., and 32-38% similarity with Arabidopsis sp. The Motif Finder server identified the protein family domain FLO_LFY and LFY_SAM, whose function is floral meristem development. Secondary structure prediction analysis indicated that the LEAFY protein belongs to the alpha (α) protein class, which is stable against mutation and thus limits structural changes in the LEAFY protein. The study findings reveal two distinct clusters of the LFY gene from the common ancestor green algae. One cluster is present in nonflowering plants that include mosses, pteridophytes, and gymnosperms, and the other cluster is present in flowering plants that include orchids, monocots, dicots, and angiosperms.

Point defects in hexagonal silicon

The importance of hexagonal Lonsdaleite silicon-germanium has been growing lately due to its possible uses in optoelectronic devices. However, very little is known about defects in the hexagonal phases of group-IV semiconductors. We extend here an efficient constrained structure prediction algorithm designed for interface reconstructions to the study of point defect geometries. With this method we perform an exhaustive structure prediction study of the most energetically favorable intrinsic defects in Lonsdaleite silicon. We obtain among the lowest-energy structures the hexagonal counterparts of all known defects of cubic silicon, together with other often more complex geometries. Neutral vacancies, fourfold-coordinated, and Frenkel defects have comparable formation energies in both hexagonal and cubic phases, while some interstitial defects become considerably more stable in the hexagonal lattice. Furthermore, due to the reduced symmetry, formation energies can depend on the orientation of the defect with respect to the $c$ axis. Finally, we calculate the density of states of the defective supercells to determine which defects lead to electronic states in the band gap, potentially affecting the performance of optoelectronic devices based on hexagonal group-IV crystals.

Structure prediction of cyclic peptides by molecular dynamics + machine learning.

Recent computational methods have made strides in discovering well-structured cyclic peptides that preferentially populate a single conformation. However, many successful cyclic-peptide therapeutics adopt multiple conformations in solution. In fact, the chameleonic properties of some cyclic peptides are likely responsible for their high cell membrane permeability. Thus, we require the ability to predict complete structural ensembles for cyclic peptides, including the majority of cyclic peptides that have broad structural ensembles, to significantly improve our ability to rationally design cyclic-peptide therapeutics. Here, we introduce the idea of using molecular dynamics simulation results to train machine learning models to enable efficient structure prediction for cyclic peptides. Using molecular dynamics simulation results for several hundred cyclic pentapeptides as the training datasets, we developed machine-learning models that can provide molecular dynamics simulation-quality predictions of structural ensembles for all the hundreds of thousands of sequences in the entire sequence space. The prediction for each individual cyclic peptide can be made using less than 1 second of computation time. Even for the most challenging classes of poorly structured cyclic peptides with broad conformational ensembles, our predictions were similar to those one would normally obtain only after running multiple days of explicit-solvent molecular dynamics simulations. The resulting method, termed StrEAMM (Structural Ensembles Achieved by Molecular Dynamics and Machine Learning), is the first technique capable of efficiently predicting complete structural ensembles of cyclic peptides without relying on additional molecular dynamics simulations, constituting a seven-order-of-magnitude improvement in speed while retaining the same accuracy as explicit-solvent simulations.

Swarm Intelligence Steers a Global Minima Search of Clusters Bound on Carbon Nanostructures

A detailed exploration of the energy landscape is a necessary criterion for an efficient structure prediction. Although computational power has increased many-folds in recent decades, a full analys...

An efficient three-Level parallel ABC algorithm for secondary structure prediction of complex RNA sequences

This paper introduces 3L-PABCfold, a parallel artificial bee colony algorithm with three-Level strategy for efficient secondary structure prediction of complex RNA sequences, by multi-processor parallel computing. Designed 3L-PABCfold is a combination of additionally proposed three sub-algorithms: three-Level parallel ABC algorithm (3L-PABC), set based ABC algorithm and discrete ABC (DABC) algorithm. Proposed 3L-PABCfold is an extension of author’s earlier developed TL-PSOfold Lalwani et al. (2016), a two-Level particle swarm optimization algorithm for RNA secondary structure prediction. 3L-PABCfold is furthermore enriched with complexity evaluation of highly obscure sequence families followed by makespan minimization and algorithm implementation in parallel computing environment.As the complex dataset, 1,00,000 RNA sequences are randomly generated with length range 139–6358 nucleotides. Level-I of the algorithm reduces the machines idle time by implementing an efficient schedule, whereas, Level-II maximizes the bonded pairs of RNA sequences and Level-III provides optimum secondary structure of RNA sequence. Moreover, three more parallel versions are developed i.e. parallel version of TL-PSOfold (TL-PPSOfold), three Level parallel version of TL-PSOfold (3L-PPSOfold) and two-Level parallel version of 3L-PABCfold (TL-PABCfold), so as to avail a fair comparison and to identify the competent strategy for the addressed problem. The performance evaluated at the criteria of sensitivity, specificity, F-measure along with the descriptive statistics and statistical significance, yields significantly better prediction accuracy for 3L-PABCfold. Further, the efficiency at the processing time criteria proves 3L-PABCfold as the most competent algorithm than TL-PPSOfold, TL-PABCfold and 3L-PPSOfold.

RNA modifications in structure prediction - Status quo and future challenges.

Chemical modifications of RNA nucleotides change their identity and characteristics and thus alter genetic and structural information encoded in the genomic DNA. tRNA and rRNA are probably the most heavily modified genes, and often depend on derivatization or isomerization of their nucleobases in order to correctly fold into their functional structures. Recent RNomics studies, however, report transcriptome wide RNA modification and suggest a more general regulation of structuredness of RNAs by this so called epitranscriptome. Modification seems to require specific substrate structures, which in turn are stabilized or destabilized and thus promote or inhibit refolding events of regulatory RNA structures. In this review, we revisit RNA modifications and the related structures from a computational point of view. We discuss known substrate structures, their properties such as sub-motifs as well as consequences of modifications on base pairing patterns and possible refolding events. Given that efficient RNA structure prediction methods for canonical base pairs have been established several decades ago, we review to what extend these methods allow the inclusion of modified nucleotides to model and study epitranscriptomic effects on RNA structures.

A Distributed Tree-based Ensemble Learning Approach for Efficient Structure Prediction of Protein

A Distributed Tree-based Ensemble Learning Approach for Efficient Structure Prediction of Protein

Unexpected Xe anions in XeLin intermetallic compounds

The reactivity of Xe is important in both fundamental chemistry and geological science. The discovery of the reductive reactivity of Xe extended the doctrinal boundary of chemistry for which a completed shell is inert to reaction. The oxidation of Xe by various elements has been explored. On the other hand, the opposite chemical inclination, i.e., gaining electrons and forming anions, has not been thoroughly studied for Xe or other noble-gas elements. In this work, we demonstrate, using first-principles calculations and an efficient structure prediction method, that Xe can form stable compounds under high pressure. These compounds are intermetallic and Xe are negatively charged. The stability of these compounds indicates that atoms or ions with completely filled shell may still gain electrons in chemical reactions.

Distributed Machine Learning Algorithms to classify Protein secondary structures for Drug Design-A Survey

The arena of secondary structure based drug design is an evolving area in recent years. Structures help us to understand how the protein functions, and will help to design chemicals to be used pharmaceutically as modifiers of enzyme activity. X-Ray structures usually provide us with a quite static picture of the protein which is considered as expensive and time consuming. The recent modern drug design and development using knowledge of proteomics must rely on computational intelligence based machine learning model structures using efficient structure prediction techniques. In recent years, Machine learning, emerging on the basis of parallel and distributed computing for handling big data, is making huge advances in many areas. In this paper, we have captured a comprehensive review of protein structure prediction methods using Machine Learning approach in Distributed environment. Overall, the results are good and depicts that the accuracy and performance of protein secondary structure prediction methods are achieved using Machine Learning techniques and these would become a powerful aid, while implemented in Distributed environment like Hadoop or Spark. This research will be helpful for the recent medicine researchers, which aids in understanding the relation between protein sequences, structure and thereby determine the function to develop various drugs and designing novel enzymes and this is considered as one of the major focused areas in recent bioinformatics research.

Lattice model for amyloid peptides: OPEP force field parametrization and applications to the nucleus size of Alzheimer's peptides.

Coarse-grained protein lattice models approximate atomistic details and keep the essential interactions. They are, therefore, suitable for capturing generic features of protein folding and amyloid formation at low computational cost. As our aim is to study the critical nucleus sizes of two experimentally well-characterized peptide fragments Aβ16-22 and Aβ37-42 of the full length Aβ1-42 Alzheimer's peptide, it is important that simulations with the lattice model reproduce all-atom simulations. In this study, we present a comprehensive force field parameterization based on the OPEP (Optimized Potential for Efficient protein structure Prediction) force field for an on-lattice protein model, which incorporates explicitly the formation of hydrogen bonds and directions of side-chains. Our bottom-up approach starts with the determination of the best lattice force parameters for the Aβ16-22 dimer by fitting its equilibrium parallel and anti-parallel β-sheet populations to all-atom simulation results. Surprisingly, the calibrated force field is transferable to the trimer of Aβ16-22 and the dimer and trimer of Aβ37-42. Encouraged by this finding, we characterized the free energy landscapes of the two decamers. The dominant structure of the Aβ16-22 decamer matches the microcrystal structure. Pushing the simulations for aggregates between 4-mer and 12-mer suggests a nucleus size for fibril formation of 10 chains. In contrast, the Aβ37-42 decamer is largely disordered with mixed by parallel and antiparallel chains, suggesting that the nucleus size is >10 peptides. Our refined force field coupled to this on-lattice model should provide useful insights into the critical nucleation number associated with neurodegenerative diseases.

Evaluation of the coarse-grained OPEP force field for protein-protein docking.

BackgroundKnowing the binding site of protein–protein complexes helps understand their function and shows possible regulation sites. The ultimate goal of protein–protein docking is the prediction of the three-dimensional structure of a protein–protein complex. Docking itself only produces plausible candidate structures, which must be ranked using scoring functions to identify the structures that are most likely to occur in nature.MethodsIn this work, we rescore rigid body protein–protein predictions using the optimized potential for efficient structure prediction (OPEP), which is a coarse-grained force field. Using a force field based on continuous functions rather than a grid-based scoring function allows the introduction of protein flexibility during the docking procedure. First, we produce protein–protein predictions using ZDOCK, and after energy minimization via OPEP we rank them using an OPEP-based soft rescoring function. We also train the rescoring function for different complex classes and demonstrate its improved performance for an independent dataset.ResultsThe trained rescoring function produces a better ranking than ZDOCK for more than 50 % of targets, rising to over 70 % when considering only enzyme/inhibitor complexes.ConclusionsThis study demonstrates for the first time that energy functions derived from the coarse-grained OPEP force field can be employed to rescore predictions for protein–protein complexes.

Ordered structures in III-Nitride ternary alloys

An efficient evolutionary structure prediction algorithm in combination with ab initio calculations is implemented in order to reveal energetically favorable superstructures of the III-Nitride ternary alloys. Several 2×2×2 32-atom supercells are used to explore the full range of concentrations, from x=0 to 1. The formation enthalpies, bandgaps, clustering and/or ordering of the atoms are investigated and the results are discussed.The formation enthalpy plots show local minima at specific concentrations, namely for x=0.25, 0.50 and 0.75, that correspond to ordered structures. The valance band maxima, conduction band minima, bandgaps and the composition-independent bowing parameters for 2nd order Vegard’s equation are calculated. The bandgap deviations from 1st order Vegard’s law show total maxima at specific concentrations. The formation enthalpies and the bandgaps cannot be accurately described by single composition-independent bowing parameters, but the bandgaps are sufficiently described by composition-dependent bowing parameters, that are established.In order to verify the rationality of the results against the size of the ab initio supercells, molecular dynamics calculations of 14×14×14 supercells of ∼104atoms using bond-order interatomic potentials are performed. The obtained local minima of the formation enthalpy for the specific alloy compositions concur with those predicted by ab initio calculations proving the results are not influenced by the supercell size.

Hydrodynamic Effects on Amyloid-β Aggregation

The self-assembly of misfolded amyloid-β (Aβ) proteins into insoluble fibrils is strongly linked to the pathogenesis of Alzheimer's disease (AD). The development of new drugs capable to inhibit fibril formation requires the understanding of the mechanism leading to the formation, in the early stages, of metastable oligomers which are the main neurotoxic species.Our investigation enlighten the role of hydrodynamic interactions (HI) in the kinetics of β-amyloidogenesis, interactions which are essential for modelling such processes occurring in highly crowed environments like cells [1].Our approach is based on a new multi-scale and multi-physics method that couples Lattice Boltzmann and Molecular Dynamics (LBMD) techniques[2,3]. In our scheme solvent mediated interactions are included naturally. For the molecular system we adopt the high resolution coarse-grained model OPEP (Optimized Potential for Efficient Protein structure prediction)[4], developed in our laboratory. For the first time, we have performed quasi all-atoms simulations for very large systems containing thousands of amyloid peptides and approaching experimental concentration. After the correct tuning of the key parameter of our coupling in order to obtain the experimental diffusivity of Aβ monomer and small oligomers, we have demonstrated that HI speed up the aggregation process. In our poster we will present a detailed characterisation of the fluctuating clusters in term of their size and the structural organisation of the peptides.