Efficient Amino Acid Conformer Search with Bayesian Optimization

Accelerated Lignocellulosic Molecule Adsorption Structure Determination

Here, we present a study combining Bayesian optimization structural inference with the machine learning interatomic potential Neural Equivariant Interatomic Potential (NequIP) to accelerate and enable the study of the adsorption of the conformationally flexible lignocellulosic molecules β-d-xylose and 1,4-β-d-xylotetraose on a copper surface. The number of structure evaluations needed to map out the relevant potential energy surfaces are reduced by Bayesian optimization, while NequIP minimizes the time spent on each evaluation, ultimately resulting in cost-efficient and reliable sampling of large systems and configurational spaces. Although the applicability of Bayesian optimization for the conformational analysis of the more flexible xylotetraose molecule is restricted by the sample complexity bottleneck, the latter can be effectively bypassed with external conformer search tools, such as the Conformer-Rotamer Ensemble Sampling Tool, facilitating the subsequent lower-dimensional global minimum adsorption structure determination. Finally, we demonstrate the applicability of the described approach to find adsorption structures practically equivalent to the density functional theory counterparts at a fraction of the computational cost.

Bayesian Optimization in Drug Discovery

Drug discovery deals with the search for initial hits and their optimization toward a targeted clinical profile. Throughout the discovery pipeline, the candidate profile will evolve, but the optimization will mainly stay a trial-and-error approach. Tons of in silico methods have been developed to improve and fasten this pipeline. Bayesian optimization (BO) is a well-known method for the determination of the global optimum of a function. In the last decade, BO has gained popularity in the early drug design phase. This chapter starts with the concept of black box optimization applied to drug design and presents some approaches to tackle it. Then it focuses on BO and explains its principle and all the algorithmic building blocks needed to implement it. This explanation aims to be accessible to people involved in drug discovery projects. A strong emphasis is made on the solutions to deal with the specific constraints of drug discovery. Finally, a large set of practical applications of BO is highlighted.

Exploring torsional conformer space with physical prior mean function-driven meta-Gaussian processes

We present a novel approach for systematically exploring the conformational space of small molecules with multiple internal torsions. Identifying unique conformers through a systematic conformational search is important for obtaining accurate thermodynamic functions (e.g., free energy), encompassing contributions from the ensemble of all local minima. Traditional geometry optimizers focus on one structure at a time, lacking transferability from the local potential-energy surface (PES) around a specific minimum to optimize other conformers. In this work, we introduce a physics-driven meta-Gaussian processes (meta-GPs) method that not only enables efficient exploration of target PES for locating local minima but, critically, incorporates physical surrogates that can be applied universally across the optimization of all conformers of the same molecule. Meta-GPs construct surrogate PESs based on the optimization history of prior conformers, dynamically selecting the most suitable prior mean function (representing prior knowledge in Bayesian learning) as a function of the optimization progress. We systematically benchmarked the performance of multiple GP variants for brute-force conformational search of amino acids. Our findings highlight the superior performance of meta-GPs in terms of efficiency, comprehensiveness of conformer discovery, and the distribution of conformers compared to conventional non-surrogate optimizers and other non-meta-GPs. Furthermore, we demonstrate that by concurrently optimizing, training GPs on the fly, and learning PESs, meta-GPs exhibit the capacity to generate high-quality PESs in the torsional space without extensive training data. This represents a promising avenue for physics-based transfer learning via meta-GPs with adaptive priors in exploring torsional conformer space.

Quantum Chemical Calculations of Flexible Tripeptide-Ni(II) Ion-Mediated Supramolecular Fragments and Comparative Analysis of Tripeptide Complexes with Various Metal(II) Ions

An automated conformational search method was employed to efficiently determine the stable conformers and weak hydrogen bonds of a flexible tripeptide coordinated with a solitary metal(II) ion in an aqueous environment. Quantum chemical calculations were performed to investigate the tendency of octahedral coordination formation between different metal(II) ions and various coordination models (ammonia molecule, chelate molecule, and flexible tripeptide). The octahedral coordination was analyzed by decomposing it into tridentate, bidentate, and monodentate coordination model complexes to assess their formation propensities and conformational properties. Additionally, population analysis, including electrostatic potential mapping and natural population analysis, was performed to identify the unique properties of the Ni(II) ion in forming octahedral coordination in crystals and to explore the potential of other metal(II) ions for self-assembling novel coordination configurations in peptide-metal compounds. Two common hydrogen bonding interactions were examined by using artificial forces to facilitate dissociation or reinforcement.

Enhancing the Search Performance of Bayesian Optimization by Creating Different Descriptor Datasets Using Density Functional Theory

Descriptors calculated from molecular structure information can be used as explanatory variables in Bayesian optimization (BO). Even though structural and descriptor information can be obtained from various databases for general compounds, information on highly confidential compounds such as pharmaceutical intermediates and active pharmaceutical ingredients cannot be retrieved from these databases. In particular, determining the stable structure and electronic state of a compound via quantum chemical calculations from descriptor information requires considerable computational time. Although descriptor information can be obtained using density functional theory (DFT), which has a relatively light computational load, only conventional combinations of basis sets and functionals can be selected before experiments instead of the best ones. Few studies have discussed these effects on the search performance of BO, and good search performance is highly dependent on the application. Therefore, we developed a method to improve the search performance of BO by using descriptors computed from several combinations of basis sets and functionals. The dataset obtained from averaging multiple descriptor sets exhibited better BO search performance than that of a single descriptor dataset. In addition, the more descriptor sets used for averaging, the better the search performance. This method has a relatively small computational load and can be easily used by those who are unfamiliar with quantum chemical calculations.

Understanding the biodegradation pathways of azo dyes by immobilized white-rot fungus, Trametes hirsuta D7, using UPLC-PDA-FTICR MS supported by in silico simulations and toxicity assessment

No biodegradation methods are absolute in the treatment of all textile dyes, which leads to structure-dependent degradation. In this study, biodegradation of three azo dyes, reactive black 5 (RB5), acid blue 113 (AB113), and acid orange 7 (AO7), was investigated using an immobilized fungus, Trametes hirsuta D7. The degraded metabolites were identified using UPLC-PDA-FTICR MS and the biodegradation pathway followed was proposed. RB5 (92%) and AB113 (97%) were effectively degraded, whereas only 30% of AO7 was degraded. Molecular docking simulations were performed to determine the reason behind the poor degradation of AO7. Weak binding affinity, deficiency in H-bonding interactions, and the absence of interactions between the azo (-NN-) group and active residues of the model laccase enzyme were responsible for the low degradation efficiency of AO7. Furthermore, cytotoxicity and genotoxicity assays confirmed that the fungus-treated dye produced non-toxic metabolites. The observations of this study will be useful for understanding and further improving enzymatic dye biodegradation.

Initial Sample Selection in Bayesian Optimization for Combinatorial Optimization of Chemical Compounds

An efficient search for optimal solutions in Bayesian optimization (BO) entails providing appropriate initial samples when building a Gaussian process regression model. For general experimental designs without compounds or molecular descriptors in explanatory variable x, selecting initial samples with a larger D-optimality allows little correlation between x in the selected samples, which leads to effective regression model building. However, in the case of experimental designs with compounds, a high correlation always exists between molecular descriptors calculated from chemical structures, and compounds with similar structures form clusters in the chemical space. Therefore, selecting the initial samples uniformly from each cluster is desirable for obtaining initial samples with maximum information on experimental conditions. As D-optimality does not work well with highly correlated molecular descriptors and does not consider information on clusters in sample selection, we propose an initial sample selection method based on clustering and apply it to the optimization of coupling reaction conditions with BO. We confirm that the proposed method reaches the optimal solution with up to 5% fewer experiments than random sampling or sampling based on D-optimality. This study makes a contribution to the initial sample selection method for BO, and we are convinced that the proposed method improves the search performance of BO in various fields of science and technology if initial samples can be determined using cluster information appropriately formed by utilizing domain knowledge.

Exploring the Conformers of an Organic Molecule on a Metal Cluster with Bayesian Optimization

Finding low-energy conformers of organic molecules is a complex problem due to the flexibilities of the molecules and the high dimensionality of the search space. When such molecules are on nanoclusters, the search complexity is exacerbated by constraints imposed by the presence of the cluster and other surrounding molecules. To address this challenge, we modified our previously developed active learning molecular conformer search method based on Bayesian optimization and density functional theory. Especially, we have developed and tested strategies to avoid steric clashes between a molecule and a cluster. In this work, we chose a cysteine molecule on a well-studied gold-thiolate cluster as a model system to test and demonstrate our method. We found that cysteine conformers in a cluster inherit the hydrogen bond types from isolated conformers. However, the energy rankings and spacings between the conformers are reordered.

Molecular Conformer Search with Low-Energy Latent Space

Identifying low-energy conformers with quantum mechanical accuracy for molecules with many degrees of freedom is challenging. In this work, we use the molecular dihedral angles as features and explore the possibility of performing molecular conformer search in a latent space with a generative model named variational auto-encoder (VAE). We bias the VAE towards low-energy molecular configurations to generate more informative data. In this way, we can effectively build a reliable energy model for the low-energy potential energy surface. After the energy model has been built, we extract local-minimum conformations and refine them with structure optimization. We have tested and benchmarked our low-energy latent-space (LOLS) structure search method on organic molecules with 5-9 searching dimensions. Our results agree with previous studies.

Protective Coating Interfaces for Perovskite Solar Cell Materials: A First-Principles Study

The protection of halide perovskites is important for the performance and stability of emergent perovskite-based optoelectronic technologies. In this work, we investigate the potential inorganic protective coating materials ZnO, SrZrO₃, and ZrO₂ for the CsPbI₃ perovskite. The optimal interface registries are identified with Bayesian optimization. We then use semilocal density functional theory (DFT) to determine the atomic structure at the interfaces of each coating material with the clean CsI-terminated surface and three reconstructed surface models with added PbI₂ and CsI complexes. For the final structures, we explore the level alignment at the interface with hybrid DFT calculations. Our analysis of the level alignment at the coating-substrate interfaces reveals no detrimental mid-gap states but rather substrate-dependent valence and conduction band offsets. While ZnO and SrZrO₃ act as insulators on CsPbI₃, ZrO₂ might be suitable as an electron transport layer with the right interface engineering.

Machine learning as a tool to engineer microstructures: Morphological prediction of tannin-based colloids using Bayesian surrogate models

ABSTRACT

Oxidized tannic acid (OTA) is a useful biomolecule with a strong tendency to form complexes with metals and proteins. In this study we open the possibility to further the application of OTA when assembled as supramolecular systems, which typically exhibit functions that correlate with shape and associated morphological features. We used machine learning (ML) to selectively engineer OTA into particles encompassing one-dimensional to three-dimensional constructs. We employed Bayesian regression to correlate colloidal suspension conditions (pH and pK _a) with the size and shape of the assembled colloidal particles. Fewer than 20 experiments were found to be sufficient to build surrogate model landscapes of OTA morphology in the experimental design space, which were chemically interpretable and endowed predictive power on data. We produced multiple property landscapes from the experimental data, helping us to infer solutions that would satisfy, simultaneously, multiple design objectives. The balance between data efficiency and the depth of information delivered by ML approaches testify to their potential to engineer particles, opening new prospects in the emerging field of particle morphogenesis, impacting bioactivity, adhesion, interfacial stabilization, and other functions inherent to OTA.

IMPACT STATEMENT

Tannic acid is a versatile bio-derived material employed in coatings, surface modifiers, and emulsion and growth stabilizers, which also imparts mild anti-viral health benefits. Our recent work on the crystallization of oxidized tannic acid (OTA) colloids opens the route toward further valuable applications, but here the functional properties tend to depend strongly on particle morphology. In this study, we eschew trial-and-error morphology exploration of OTA particles in favor of a data-driven approach. We digitalized the experimental observations and input them into a Gaussian process regression algorithm to generate morphology surrogate models. These help us to visualize particle morphology in the design space of material processing conditions, and thus determine how to selectively engineer one-dimensional or three-dimensional particles with targeted functionalities. We extend this approach to visualize other experimental outcomes, including particle yield and particle surface-to-volume ratio, which are useful for the design of products based on OTA particles. Our findings demonstrate the use of data-efficient surrogate models for general materials engineering purposes and facilitate the development of next-generation OTA-based applications.

SUPPLEMENTARY INFORMATION

The online version contains supplementary material available at 10.1557/s43577-021-00183-4.

Autonomous Reaction Network Exploration in Homogeneous and Heterogeneous Catalysis

Autonomous computations that rely on automated reaction network elucidation algorithms may pave the way to make computational catalysis on a par with experimental research in the field. Several advantages of this approach are key to catalysis: (i) automation allows one to consider orders of magnitude more structures in a systematic and open-ended fashion than what would be accessible by manual inspection. Eventually, full resolution in terms of structural varieties and conformations as well as with respect to the type and number of potentially important elementary reaction steps (including decomposition reactions that determine turnover numbers) may be achieved. (ii) Fast electronic structure methods with uncertainty quantification warrant high efficiency and reliability in order to not only deliver results quickly, but also to allow for predictive work. (iii) A high degree of autonomy reduces the amount of manual human work, processing errors, and human bias. Although being inherently unbiased, it is still steerable with respect to specific regions of an emerging network and with respect to the addition of new reactant species. This allows for a high fidelity of the formalization of some catalytic process and for surprising in silico discoveries. In this work, we first review the state of the art in computational catalysis to embed autonomous explorations into the general field from which it draws its ingredients. We then elaborate on the specific conceptual issues that arise in the context of autonomous computational procedures, some of which we discuss at an example catalytic system.

GRAPHICAL ABSTRACT

SUPPLEMENTARY INFORMATION

The online version contains supplementary material available at 10.1007/s11244-021-01543-9.

TorsiFlex: an automatic generator of torsional conformers. Application to the twenty proteinogenic amino acids

In this work, we introduce TorsiFlex, a user-friendly software written in Python 3 and designed to find all the torsional conformers of flexible acyclic molecules in an automatic fashion. For the mapping of the torsional potential energy surface, the algorithm implemented in TorsiFlex combines two searching strategies: preconditioned and stochastic. The former is a type of systematic search based on chemical knowledge and should be carried out before the stochastic (random) search. The algorithm applies several validation tests to accelerate the exploration of the torsional space. For instance, the optimized structures are stored and this information is used to prevent revisiting these points and their surroundings in future iterations. TorsiFlex operates with a dual-level strategy by which the initial search is carried out at an inexpensive electronic structure level of theory and the located conformers are reoptimized at a higher level. Additionally, the program takes advantage of conformational enantiomerism, when possible. As a case study, and in order to exemplify the effectiveness and capabilities of this program, we have employed TorsiFlex to locate the conformers of the twenty proteinogenic amino acids in their neutral canonical form. TorsiFlex has produced a number of conformers that roughly doubles the amount of the most complete work to date.

Effective Use of Empirical Data for Virtual Screening against APJR GPCR Receptor

Alzheimer's disease is a neurodegenerative disorder incompatible with normal daily activity, affecting one in nine people. One of its potential targets is the apelin receptor (APJR), a G-protein coupled receptor, which presents considerably high expression levels in the central nervous system. In silico studies of APJR drug-like molecule binding are in small numbers while high throughput screenings (HTS) are already sufficiently many to devise efficient drug design strategies. This presents itself as an opportunity to optimize different steps in future large scale virtual screening endeavours. Here, we ran a first stage docking simulation against a library of 95 known binders and 3829 generated decoys in an effort to improve the rescoring stage. We then analyzed receptor binding site structure and ligands binding poses to describe their interactions. As a result, we devised a simple and straightforward virtual screening Stage II filtering score based on search space extension followed by a geometric estimation of the ligand-binding site fitness. Having this score, we used an ensemble of receptors generated by Hamiltonian Monte Carlo simulation and reported the results. The improvements shown herein prove that our ensemble docking protocol is suited for APJR and can be easily extrapolated to other GPCRs.

Integration of theory, simulation, artificial intelligence and virtual reality: a four-pillar approach for reconciling accuracy and interpretability in computational spectroscopy

The established pillars of computational spectroscopy are theory and computer based simulations. Recently, artificial intelligence and virtual reality are becoming the third and fourth pillars of an integrated strategy for the investigation of complex phenomena. The main goal of the present contribution is the description of some new perspectives for computational spectroscopy, in the framework of a strategy in which computational methodologies at the state of the art, high-performance computing, artificial intelligence and virtual reality tools are integrated with the aim of improving research throughput and achieving goals otherwise not possible. Some of the key tools (e.g., continuous molecular perception model and virtual multifrequency spectrometer) and theoretical developments (e.g., non-periodic boundaries, joint variational-perturbative models) are shortly sketched and their application illustrated by means of representative case studies taken from recent work by the authors. Some of the results presented are already well beyond the state of the art in the field of computational spectroscopy, thereby also providing a proof of concept for other research fields.

Efficient Search for Energetically Favorable Molecular Conformations against Metastable States via Gray-Box Optimization

In order to accurately understand and estimate molecular properties, finding energetically favorable molecular conformations is the most fundamental task for atomistic computational research on molecules and materials. Geometry optimization based on quantum chemical calculations has enabled the conformation prediction of arbitrary molecules, including de novo ones. However, it is computationally expensive to perform geometry optimizations for enormous conformers. In this study, we introduce the gray-box optimization (GBO) framework, which enables optimal control over the entire geometry optimization process, among multiple conformers. Algorithms designed for GBO roughly estimate energetically preferable conformers during their geometry optimization iterations. They then preferentially compute promising conformers. To evaluate the performance of the GBO framework, we applied it to a test set consisting of seven dipeptides and mycophenolic acid to determine their stable conformations at the density functional theory level. We thus preferentially obtained energetically favorable conformations. Furthermore, the computational costs required to find the most stable conformation were significantly reduced (approximately 1% on average, compared to the naive approach for the dipeptides).