Initial Sample Selection in Bayesian Optimization for Combinatorial Optimization of Chemical Compounds
Ontology highlight
ABSTRACT:
SUBMITTER:
PROVIDER: S-EPMC9850731 | biostudies-literature | 2022 Dec
REPOSITORIES: biostudies-literature
Similar Datasets
Project description:Antibodies are multimeric proteins capable of highly specific molecular recognition. The complementarity determining region 3 of the antibody variable heavy chain (CDRH3) often dominates antigen-binding specificity. Hence, it is a priority to design optimal antigen-specific CDRH3 to develop therapeutic antibodies. The combinatorial structure of CDRH3 sequences makes it impossible to query binding-affinity oracles exhaustively. Moreover, antibodies are expected to have high target specificity and developability. Here, we present AntBO, a combinatorial Bayesian optimization framework utilizing a CDRH3 trust region for an in silico design of antibodies with favorable developability scores. The in silico experiments on 159 antigens demonstrate that AntBO is a step toward practically viable in vitro antibody design. In under 200 calls to the oracle, AntBO suggests antibodies outperforming the best binding sequence from 6.9 million experimentally obtained CDRH3s. Additionally, AntBO finds very-high-affinity CDRH3 in only 38 protein designs while requiring no domain knowledge.
| S-EPMC9939385 | biostudies-literature
Project description: Not available
| S-EPMC10873405 | biostudies-literature
Project description:Electron ptychography provides new opportunities to resolve atomic structures with deep sub-angstrom spatial resolution and to study electron-beam sensitive materials with high dose efficiency. In practice, obtaining accurate ptychography images requires simultaneously optimizing multiple parameters that are often selected based on trial-and-error, resulting in low-throughput experiments and preventing wider adoption. Here, we develop an automatic parameter selection framework to circumvent this problem using Bayesian optimization with Gaussian processes. With minimal prior knowledge, the workflow efficiently produces ptychographic reconstructions that are superior to those processed by experienced experts. The method also facilitates better experimental designs by exploring optimized experimental parameters from simulated data.
| S-EPMC9296498 | biostudies-literature
Project description:Automatic Chemical Design is a framework for generating novel molecules with optimized properties. The original scheme, featuring Bayesian optimization over the latent space of a variational autoencoder, suffers from the pathology that it tends to produce invalid molecular structures. First, we demonstrate empirically that this pathology arises when the Bayesian optimization scheme queries latent space points far away from the data on which the variational autoencoder has been trained. Secondly, by reformulating the search procedure as a constrained Bayesian optimization problem, we show that the effects of this pathology can be mitigated, yielding marked improvements in the validity of the generated molecules. We posit that constrained Bayesian optimization is a good approach for solving this kind of training set mismatch in many generative tasks involving Bayesian optimization over the latent space of a variational autoencoder.
| S-EPMC7067240 | biostudies-literature
Project description:Protein engineering by combinatorial site-directed mutagenesis evaluates a portion of the sequence space near a target protein, seeking variants with improved properties (e.g., stability, activity, immunogenicity). In order to improve the hit-rate of beneficial variants in such mutagenesis libraries, we develop methods to select optimal positions and corresponding sets of the mutations that will be used, in all combinations, in constructing a library for experimental evaluation. Our approach, OCoM (Optimization of Combinatorial Mutagenesis), encompasses both degenerate oligonucleotides and specified point mutations, and can be directed accordingly by requirements of experimental cost and library size. It evaluates the quality of the resulting library by one- and two-body sequence potentials, averaged over the variants. To ensure that it is not simply recapitulating extant sequences, it balances the quality of a library with an explicit evaluation of the novelty of its members. We show that, despite dealing with a combinatorial set of variants, in our approach the resulting library optimization problem is actually isomorphic to single-variant optimization. By the same token, this means that the two-body sequence potential results in an NP-hard optimization problem. We present an efficient dynamic programming algorithm for the one-body case and a practically-efficient integer programming approach for the general two-body case. We demonstrate the effectiveness of our approach in designing libraries for three different case study proteins targeted by previous combinatorial libraries--a green fluorescent protein, a cytochrome P450, and a beta lactamase. We found that OCoM worked quite efficiently in practice, requiring only 1 hour even for the massive design problem of selecting 18 mutations to generate 10⁷ variants of a 443-residue P450. We demonstrate the general ability of OCoM in enabling the protein engineer to explore and evaluate trade-offs between quality and novelty as well as library construction technique, and identify optimal libraries for experimental evaluation.
| S-EPMC5220575 | biostudies-literature
Project description:Mutations in the BRCA1 and BRCA2 genes are known to be highly associated with breast cancer. Identifying both shared and unique transcript expression patterns in blood samples from these groups can shed insight into if and how the disease mechanisms differ among individuals by mutation status, but this is challenging in the high-dimensional setting. A recent method, Bayesian Multi-Study Factor Analysis (BMSFA), identifies latent factors common to all studies (or equivalently, groups) and latent factors specific to individual studies. However, BMSFA does not allow for factors shared by more than one but less than all studies. This is critical in our context, as we may expect some but not all signals to be shared by BRCA1-and BRCA2-mutation carriers but not necessarily other high-risk groups. We extend BMSFA by introducing a new method, Tetris, for Bayesian combinatorial multi-study factor analysis, which identifies latent factors that any combination of studies or groups can share. We model the subsets of studies that share latent factors with an Indian Buffet Process, and offer a way to summarize uncertainty in the sharing patterns using credible balls. We test our method with an extensive range of simulations, and showcase its utility not only in dimension reduction but also in covariance estimation. When applied to transcript expression data from high-risk families grouped by mutation status, Tetris reveals the features and pathways characterizing each group and the sharing patterns among them. Finally, we further extend Tetris to discover groupings of samples when group labels are not provided, which can elucidate additional structure in these data.
| S-EPMC10543692 | biostudies-literature
Project description:Response surface methodology (RSM) with a Box-Behnken design (BBD) was used to optimize the extraction of bioactive compounds from Ephedra fragilis. The results suggested that extraction with 61.93% ethanol at 44.43 °C for 15.84 h was the best solution for this combination of variables. The crude ethanol extract (CEE) obtained under optimum extraction conditions was sequentially fractionated with solvents of increasing polarity. The content of total phenolic (TP) and total flavonoid (TF) as well as the antioxidant and antiglycation activities were measured. The phytochemical fingerprint profile of the fraction with the highest activity was characterized by using RP-HPLC. The ethyl acetate fraction (EAF) had the highest TP and TF contents and exhibited the most potent antioxidant and antiglycation activities. The Pearson correlation analysis results showed that TP and TF contents were highly significantly correlated with the antioxidant and antiglycation activities. Totally, six compounds were identified in the EAF of E. fragilis, including four phenolic acids and two flavonoids. Additionally, molecular docking analysis also showed the possible connection between identified bioactive compounds and their mechanisms of action. Our results suggest new evidence on the antioxidant and antiglycation activities of E. fragilis bioactive compounds that may be applied in the treatment and prevention of aging and glycation-associated complications.
| S-EPMC8512229 | biostudies-literature
Project description:The development of spintronic devices demands the existence of materials with some kind of spin splitting (SS). In this Data Descriptor, we build a database of ab initio calculated SS in 2D materials. More than that, we propose a workflow for materials design integrating an inverse design approach and a Bayesian inference optimization. We use the prediction of SS prototypes for spintronic applications as an illustrative example of the proposed workflow. The prediction process starts with the establishment of the design principles (the physical mechanism behind the target properties), that are used as filters for materials screening, and followed by density functional theory (DFT) calculations. Applying this process to the C2DB database, we identify and classify 358 2D materials according to SS type at the valence and/or conduction bands. The Bayesian optimization captures trends that are used for the rationalized design of 2D materials with the ideal conditions of band gap and SS for potential spintronics applications. Our workflow can be applied to any other material property.
| S-EPMC9054849 | biostudies-literature
Project description:Traditional optimization methods using one variable at a time approach waste time and chemicals and assume that different parameters are independent from one another. Hence, a simpler, more practical, and rapid process for predicting reaction conditions that can be applied to several manufacturing environmentally sustainable processes is highly desirable. In this study, biaryl compounds were synthesized efficiently using an organic Brønsted acid catalyst in a flow system. Bayesian optimization-assisted multi-parameter screening, which employs one-hot encoding and appropriate acquisition function, rapidly predicted the suitable conditions for the synthesis of 2-amino-2'-hydroxy-biaryls (maximum yield of 96%). The established protocol was also applied in an optimization process for the efficient synthesis of 2,2'-dihydroxy biaryls (up to 97% yield). The optimized reaction conditions were successfully applied to gram-scale synthesis. We believe our algorithm can be beneficial as it can screen a reactor design without complicated quantification and descriptors.
| S-EPMC9814103 | biostudies-literature
Project description:BackgroundThe joint study of multiple datasets has become a common technique for increasing statistical power in detecting biomarkers obtained from smaller studies. The approach generally followed is based on the fact that as the total number of samples increases, we expect to have greater power to detect associations of interest. This methodology has been applied to genome-wide association and transcriptomic studies due to the availability of datasets in the public domain. While this approach is well established in biostatistics, the introduction of new combinatorial optimization models to address this issue has not been explored in depth. In this study, we introduce a new model for the integration of multiple datasets and we show its application in transcriptomics.MethodsWe propose a new combinatorial optimization problem that addresses the core issue of biomarker detection in integrated datasets. Optimal solutions for this model deliver a feature selection from a panel of prospective biomarkers. The model we propose is a generalised version of the (α,β)-k-Feature Set problem. We illustrate the performance of this new methodology via a challenging meta-analysis task involving six prostate cancer microarray datasets. The results are then compared to the popular RankProd meta-analysis tool and to what can be obtained by analysing the individual datasets by statistical and combinatorial methods alone.ResultsApplication of the integrated method resulted in a more informative signature than the rank-based meta-analysis or individual dataset results, and overcomes problems arising from real world datasets. The set of genes identified is highly significant in the context of prostate cancer. The method used does not rely on homogenisation or transformation of values to a common scale, and at the same time is able to capture markers associated with subgroups of the disease.
| S-EPMC4480358 | biostudies-literature