QM assisted ML for 19F NMR chemical shift prediction.

BACKGROUND: Ligand-observed 19F NMR detection is an efficient method for screening libraries of fluorinated molecules in fragment-based drug design campaigns. Screening fluorinated molecules in large mixtures makes 19F NMR a high-throughput method. Typically, these mixtures are generated from pools of well-characterized fragments. By predicting 19F NMR chemical shift, mixtures could be generated for arbitrary fluorinated molecules facilitating for example focused screens. METHODS: In a previous publication, we introduced a method to predict 19F NMR chemical shift using rooted fluorine fingerprints and machine learning (ML) methods. Having observed that the quality of the prediction depends on similarity to the training set, we here propose to assist the prediction with quantum mechanics (QM) based methods in cases where compounds are not well covered by a training set. RESULTS: Beyond similarity, the performance of ML methods could be associated with individual features in compounds. A combination of both could be used as a procedure to split input data sets into those that could be predicted by ML and those that required QM processing. We could show on a proprietary fluorinated fragment library, known as LEF (Local Environment of Fluorine), and a public Enamine data set of 19F NMR chemical shifts that ML and QM methods could synergize to outperform either method individually. Models built on Enamine data, as well as model building and QM workflow tools, can be found at https://github.com/PatrickPenner/lefshift and https://github.com/PatrickPenner/lefqm .

FastGrow: on-the-fly growing and its application to DYRK1A.

Fragment-based drug design is an established routine approach in both experimental and computational spheres. Growing fragment hits into viable ligands has increasingly shifted into the spotlight. FastGrow is an application based on a shape search algorithm that addresses this challenge at high speeds of a few milliseconds per fragment. It further features a pharmacophoric interaction description, ensemble flexibility, as well as geometry optimization to become a fully fledged structure-based modeling tool. All features were evaluated in detail on a previously reported collection of fragment growing scenarios extracted from crystallographic data. FastGrow was also shown to perform competitively versus established docking software. A case study on the DYRK1A kinase, using recently reported new chemotypes, illustrates FastGrow's features in practice and its ability to identify active fragments. FastGrow is freely available to the public as a web server at https://fastgrow.plus/ and is part of the SeeSAR 3D software package.

ProteinsPlus: a comprehensive collection of web-based molecular modeling tools.

Upon the ever-increasing number of publicly available experimentally determined and predicted protein and nucleic acid structures, the demand for easy-to-use tools to investigate these structural models is higher than ever before. The ProteinsPlus web server (https://proteins.plus) comprises a growing collection of molecular modeling tools focusing on protein-ligand interactions. It enables quick access to structural investigations ranging from structure analytics and search methods to molecular docking. It is by now well-established in the community and constantly extended. The server gives easy access not only to experts but also to students and occasional users from the field of life sciences. Here, we describe its recently added new features and tools, beyond them a novel method for on-the-fly molecular docking and a search method for single-residue substitutions in local regions of a protein structure throughout the whole Protein Data Bank. Finally, we provide a glimpse into new avenues for the annotation of AlphaFold structures which are directly accessible via a RESTful service on the ProteinsPlus web server.

The Torsion Library: Semiautomated Improvement of Torsion Rules with SMARTScompare.

The Torsion Library is a collection of torsion motifs associated with angle distributions, derived from crystallographic databases. It is used in strain assessment, conformer generation, and geometry optimization. A hierarchical structure of expert curated SMARTS defines the chemical environments of rotatable bonds and associates these with preferred angles. SMARTS can be very complex and full of implications, which make them difficult to maintain manually. Recent developments in automatically comparing SMARTS patterns can be applied to the Torsion Library to ensure its correctness. We specifically discuss the implementation and the limits of such a procedure in the context of torsion motifs and show several examples of how the Torsion Library benefits from this. All automated changes are validated manually and then shown to have an effect on the angle distributions by correcting matching behavior. The corrected Torsion Library itself is available including both PDB as well as CSD histograms in the Supporting Information and can be used to evaluate rotatable bonds at https://torsions.zbh.uni-hamburg.de.

Comparison of Combinatorial Fragment Spaces and Its Application to Ultralarge Make-on-Demand Compound Catalogs.

The set of chemical compounds shared by two or more chemical libraries is assessed routinely as means of comparing these libraries for various applications. Traditionally this is achieved by comparing the members of the chemical libraries individually for identity. This approach becomes impractical when operating on chemical libraries exceeding billions or even trillions of compounds in size. As a result, no such analysis exists for ultralarge chemical spaces like the Enamine REAL Space containing over 20 billion compounds. In this work, we present a novel tool called SpaceCompare for the overlap calculation of large, nonenumerable combinatorial fragment spaces. In contrast to existing methods, SpaceCompare utilizes topological fingerprints and the combinatorial character of these chemical spaces. The tool is able to determine the exact overlap of prominent spaces like Enamine's REAL Space, WuXi's GalaXi Space, and Otava's CHEMriya for the first time.

Topological Similarity Search in Large Combinatorial Fragment Spaces.

In similarity-driven virtual screening, molecular fingerprints are widely used to assess the similarity of all compounds contained in a chemical library to a query compound of interest. This similarity analysis is traditionally done for each member of the library individually. When encoding chemical spaces that surpass billions of compounds in size, it becomes impractical to enumerate all their products, let alone assess their similarity, deeming this approach impossible without investing a substantial amount of resources. In this work, we present a novel search algorithm named SpaceLight for topological fingerprint similarity searching in large, practically non-enumerable combinatorial fragment spaces. In contrast to existing methods, SpaceLight is able to utilize the combinatorial character of these chemical spaces for efficiency while maintaining a high correlation of the description of molecular similarity to well-known molecular fingerprints like ECFP. The resulting software is able to search prominent spaces like EnamineREAL with more than 10 billion compounds in seconds on a standard desktop computer.

A Consistent Scheme for Gradient-Based Optimization of Protein-Ligand Poses.

Scoring and numerical optimization of protein-ligand poses is an integral part of docking tools. Although many scoring functions exist, many of them are not continuously differentiable and they are rarely explicitly analyzed with respect to their numerical optimization behavior. Here, we present a consistent scheme for pose scoring and gradient-based pose optimization. It consists of a novel variant of the BFGS algorithm enabling step-length control, named LSL-BFGS (limited step length BFGS), and the empirical JAMDA scoring function designed for pose prediction and good numerical optimizability. The JAMDA scoring function shows a high pose prediction performance in the CASF-2016 docking power benchmark, top-ranking a pose with an RMSD of ≤2 Å in about 89% of the cases. The combination of JAMDA scoring with the LSL-BFGS algorithm shows a significantly higher optimization locality (i.e., no excessive movement of poses) than with the classical BFGS algorithm while retaining the characteristically low number of scoring function evaluations. The JAMDA scoring and optimization scheme is freely available for noncommercial use and academic research.

Shape-Based Descriptors for Efficient Structure-Based Fragment Growing.

Structure-based fragment growing is one of the key techniques in fragment-based drug design. Fragment growing is commonly practiced based on structural and biophysical data. Computational workflows are employed to predict which fragment elaborations could lead to high-affinity binders. Several such workflows exist but many are designed to be long running noninteractive systems. Shape-based descriptors have been proven to be fast and perform well at virtual-screening tasks. They could, therefore, be applied to the fragment-growing problem to enable an interactive fragment-growing workflow. In this work, we describe and analyze the use of specific shape-based directional descriptors for the task of fragment growing. The performance of these descriptors that we call ray volume matrices (RVMs) is evaluated on two data sets containing protein-ligand complexes. While the first set focuses on self-growing, the second measures practical performance in a cross-growing scenario. The runtime of screenings using RVMs as well as their robustness to three dimensional perturbations is also investigated. Overall, it can be shown that RVMs are useful to prefilter fragment candidates. For up to 84% of the 3299 generated self-growing cases and for up to 66% of the 326 generated cross-growing cases, RVMs could create poses with less than 2 Å root-mean-square deviation to the crystal structure with average query speeds of around 30,000 conformations per second. This opens the door for fast explorative screenings of fragment libraries.

Connected Subgraph Fingerprints: Representing Molecules Using Exhaustive Subgraph Enumeration.

Molecular fingerprints are an efficient and widely used method for similarity-driven virtual screening. Most fingerprint methods can be distinguished by the class of structural features considered. The Connected Subgraph Fingerprint (CSFP) overcomes this limitation and regards all structural features of a compound. This results in a more complete feature space and high adaptive potential to certain application scenarios. The novel descriptor surpasses widely used fingerprint methods in some cases and opens the way for topological search in combinatorial fragment spaces.