Review of ring perception algorithms for chemical graphs

D3Rings: A Fast and Accurate Method for Ring System Identification and Deep Generation of Drug-Like Cyclic Compounds

Continuous exploration of the chemical space of molecules to find ligands with high affinity and specificity for specific targets is an important topic in drug discovery. A focus on cyclic compounds, particularly natural compounds with diverse scaffolds, provides important insights into novel molecular structures for drug design. However, the complexity of their ring structures has hindered the applicability of widely accepted methods and software for the systematic identification and classification of cyclic compounds. Herein, we successfully developed a new method, D3Rings, to identify acyclic, monocyclic, spiro ring, fused and bridged ring, and cage ring compounds, as well as macrocyclic compounds. By using D3Rings, we completed the statistics of cyclic compounds in three different databases, e.g., ChEMBL, DrugBank, and COCONUT. The results demonstrated the richness of ring structures in natural products, especially spiro, macrocycles, and fused and bridged rings. Based on this, three deep generative models, namely, VAE, AAE, and CharRNN, were trained and used to construct two data sets similar to DrugBank and COCONUT but 10 times larger than them. The enlarged data sets were then used to explore the molecular chemical space, focusing on complex ring structures, for novel drug discovery and development. Docking experiments with the newly generated COCONUT-like data set against three SARS-CoV-2 target proteins revealed that an expanded compound database improves molecular docking results. Cyclic structures exhibited the best docking scores among the top-ranked docking molecules. These results suggest the importance of exploring the chemical space of structurally novel cyclic compounds and continuous expansion of the library of drug-like compounds to facilitate the discovery of potent ligands with high binding affinity to specific targets. D3Rings is now freely available at http://www.d3pharma.com/D3Rings/.

A molecule perturbation software library and its application to study the effects of molecular design constraints

Computational molecular design can yield chemically unreasonable compounds when performed carelessly. A popular strategy to mitigate this risk is mimicking reference chemistry. This is commonly achieved by restricting the way in which molecules are constructed or modified. While it is well established that such an approach helps in designing chemically appealing molecules, concerns about these restrictions impacting chemical space exploration negatively linger. In this work we present a software library for constrained graph-based molecule manipulation and showcase its functionality by developing a molecule generator. Said generator designs molecules mimicking reference chemical features of differing granularity. We find that restricting molecular construction lightly, beyond the usual positive effects on drug-likeness and synthesizability of designed molecules, provides guidance to optimization algorithms navigating chemical space. Nonetheless, restricting molecular construction excessively can indeed hinder effective chemical space exploration.

Iterative Cup Overlapping: An Efficient Identification Algorithm for Cage Structures of Amorphous Phase Hydrates

Molecular dynamics studies have revealed that the nucleation pathway of clathrate hydrates involves the evolution from amorphous to crystalline hydrates. In this study, complete cages are further classified into the standard edge-saturated cages (SECs) and nonstandard edge-saturated cages (non-SECs). Centered on studying the structure and evolution of non-SECs and SECs, we propose a novel and efficient algorithm, iterative cup overlapping (ICO), to monitor hydrate nucleation and growth in molecular simulations by identifying SECs and discuss possible causes of the instability of non-SECs. Manipulation of topological information makes it possible for ICO to avoid the repeated searches for identified cages and deduce all SECs with low time costs, improving the efficiency of identification significantly. The accuracy and efficiency of ICO were verified by comparing the identification results with other well-proven algorithms. Furthermore, it was found that non-SECs have short lifetimes and eventually decompose or reorganize into more stable structures. Some evidence suggests that the instability of non-SECs is closely related to the hydrogen-bonding configuration of water-ring aggregations that they contain. The spontaneous evolution of the hydrogen-bonding network into the tetrahedral network may be the main factor that causes the conversion of QWRAs and the evolution of non-SECs.

A nonparametric weighted feature extraction-based method for c-Jun N-terminal kinase-3 inhibitor prediction

In this work, the application of a new strategy called NWFE ensemble (nonparametric weighted feature extraction ensemble) method is proposed. Subspace-supervised projections based on NWFE are incorporated into the construction of ensembles of classifiers to facilitate the correct classification of wrongly classified instances without being detrimental to the overall performance of the ensemble. The performance of NWFE is investigated with a c-Jun N-terminal kinase-3 inhibitor benchmark dataset using different chemical compound representation models. Compared with the standard method, the results obtained show that the applied method improves the prediction performance using two classifiers based on decision trees and support vector machines.

Molecular activity prediction by means of supervised subspace projection based ensembles of classifiers

This paper proposes a method for molecular activity prediction in QSAR studies using ensembles of classifiers constructed by means of two supervised subspace projection methods, namely nonparametric discriminant analysis (NDA) and hybrid discriminant analysis (HDA). We studied the performance of the proposed ensembles compared to classical ensemble methods using four molecular datasets and eight different models for the representation of the molecular structure. Using several measures and statistical tests for classifier comparison, we observe that our proposal improves the classification results with respect to classical ensemble methods. Therefore, we show that ensembles constructed using supervised subspace projections offer an effective way of creating classifiers in cheminformatics.

AceDRG: a stereochemical description generator for ligands

The program AceDRG is designed for the derivation of stereochemical information about small molecules. It uses local chemical and topological environment-based atom typing to derive and organize bond lengths and angles from a small-molecule database: the Crystallography Open Database (COD). Information about the hybridization states of atoms, whether they belong to small rings (up to seven-membered rings), ring aromaticity and nearest-neighbour information is encoded in the atom types. All atoms from the COD have been classified according to the generated atom types. All bonds and angles have also been classified according to the atom types and, in a certain sense, bond types. Derived data are tabulated in a machine-readable form that is freely available from CCP4. AceDRG can also generate stereochemical information, provided that the basic bonding pattern of a ligand is known. The basic bonding pattern is perceived from one of the computational chemistry file formats, including SMILES, mmCIF, SDF MOL and SYBYL MOL2 files. Using the bonding chemistry, atom types, and bond and angle tables generated from the COD, AceDRG derives the `ideal' bond lengths, angles, plane groups, aromatic rings and chirality information, and writes them to an mmCIF file that can be used by the refinement program REFMAC5 and the model-building program Coot. Other refinement and model-building programs such as PHENIX and BUSTER can also use these files. AceDRG also generates one or more coordinate sets corresponding to the most favourable conformation(s) of a given ligand. AceDRG employs RDKit for chemistry perception and for initial conformation generation, as well as for the interpretation of SMILES strings, SDF MOL and SYBYL MOL2 files.

Efficient ring perception for the Chemistry Development Kit

BACKGROUND

The Chemistry Development Kit (CDK) is an open source Java library for manipulating and processing chemical information. A key aspect in handling chemical structures is the determination of the chemical rings. The rings of a structure are used areas including descriptors, stereochemistry, similarity, screening and atom typing. The CDK includes multiple algorithms for determining the rings of a structure on demand. Non-unique descriptions of rings were often used due to the slower performance of the unique alternatives.

RESULTS

Efficient algorithms for handling chemical ring perception have been implemented and optimised in the CDK. The algorithms provide much faster computation of new and existing types of rings. Several optimisation and implementation considerations are discussed which improve real case usage. The performance is measured on several publicly available data sets and in several cases the new implementations were found to be more than an order of magnitude faster.

CONCLUSIONS

Algorithmic improvements allow handling of much larger datasets in reasonable time. Faster computation allows more appropriate rings to be utilised in procedures such as aromaticity. Several areas that require ring perception have also seen a noticeable improvement. The time taken to compute the unique rings is now comparable allowing a correct usage throughout the toolkit. All source code is open source and freely available.

MATCH: an atom-typing toolset for molecular mechanics force fields

We introduce a toolset of program libraries collectively titled multipurpose atom-typer for CHARMM (MATCH) for the automated assignment of atom types and force field parameters for molecular mechanics simulation of organic molecules. The toolset includes utilities for the conversion of multiple chemical structure file formats into a molecular graph. A general chemical pattern-matching engine using this graph has been implemented whereby assignment of molecular mechanics atom types, charges, and force field parameters are achieved by comparison against a customizable list of chemical fragments. While initially designed to complement the CHARMM simulation package and force fields by generating the necessary input topology and atom-type data files, MATCH can be expanded to any force field and program, and has core functionality that makes it extendable to other applications such as fragment-based property prediction. In this work, we demonstrate the accurate construction of atomic parameters of molecules within each force field included in CHARMM36 through exhaustive cross validation studies illustrating that bond charge increment rules derived from one force field can be transferred to another. In addition, using leave-one-out substitution it is shown that it is also possible to substitute missing intra and intermolecular parameters with ones included in a force field to complete the parameterization of novel molecules. Finally, to demonstrate the robustness of MATCH and the coverage of chemical space offered by the recent CHARMM general force field (Vanommeslaeghe, et al., J Comput Chem 2010, 31, 671), one million molecules from the PubChem database of small molecules are typed, parameterized, and minimized.

Hydrogen bonding and molecular aggregates in liquid methanol, ethanol, and 1-propanol

We present a detailed and comprehensive structural study of molecular models of liquid methanol, ethanol, and 1-propanol that originate from a series of reverse Monte Carlo (RMC), molecular dynamics (MD), and united-atom Monte Carlo (UA:MC) simulations. We compare several modeling approaches: RMC simulations that employ experimental neutron and X-ray diffraction data as sole constraints, RMC with diffraction data complemented with partial radial distribution functions (PRDFs) from MD or UA:MC, and conventional MD and UA:MC simulations. The assessment is done in view of the structural parameters of the hydrogen bond and resulting morphological characteristics of molecular aggregates. To achieve these tasks, a computer program for structural analysis of molecular configurations together with the appropriate aggregate classification scheme has been developed. We have analyzed the morphology of clusters, their probability, and size distributions. Any cyclic structures that appeared in the configurations were extracted and characterized in the same manner. We found that MD and UA:MC simulations resulted in configurations with bulkier, more threadlike aggregates that were not entirely consistent with the experimental evidence from diffraction experiments. A combination of neutron and X-ray diffraction data with PRDFs from MD simulations, simultaneously applied as constraints in the RMC procedure, proved to be a modeling approach with the most conclusive results.

A robust method for searching the smallest set of smallest rings with a path-included distance matrix

The perception of rings in graphs is widely used in many fields of science and engineering. Algorithms developed in the chemistry community, called smallest set of smallest rings (SSSR), applicable only for simple graphs or chemical structures. In contrast, algorithms developed by the computer science community, called minimum cycle basis (MCB) are identical to SSSR yet exhibit greater robustness. MCB-based algorithms can correctly reveal all rings in any complex graph. However, they are slow when applied to large complex graphs due to the inherent limitations of the algorithms used. Here, we suggest a heuristic method called RP-Path. This method is a robust, simple, and fast search method with O(n(3)) runtime algorithm that correctly identifies the SSSR of all of the test case of complex graphs by using approach different from the MCB-based method. Both the robustness and improvement in speed are achieved by using a path-included distance matrix and describing the characteristic features of rings in the matrix. This method is accurate and faster than any other methods and may find many application in various fields of science and engineering that use complicated graphs with thousands of nodes.

Perspective Assessment of COX-1 and COX-2 Selectivity of Nonsteroidal Anti-Inflammatory Drugs from Clinical Practice: Use of Genetic Function Approximation

The beneficial action of nonsteroidal anti-inflammatory drugs (NSAIDs) is associated with the inhibition of cyclooxygenase-2 (COX-2), whereas their harmful side effects are associated with the inhibition of COX-1. In order to understand a meaningful comparison of both classical NSAIDs and newer COX-2 drugs, a series of molecules from varied classes of COX-2 inhibitors was studied by the application of three-dimensional quantitative structure-activity relationships (3D-QSAR) using molecular descriptors obtained by genetic function approximation. The features responsible for the dual inhibition of COX-1 and COX-2 and the selective inhibition of COX-2 with factors contributing to the maintenance of optimum selectivity were identified. The QSAR models revealed the importance of thermodynamic, electronic, structural, and molecular shape analysis parameters, which can reasonably modulate the selectivity pattern to avoid unsolicited side effects. An improved understanding to rationalize the COX-1 and COX-2 binding profiles could be gained to develop safe drug design methods.

Complexity of chemical graphs in terms of size, branching, and cyclicity

Chemical graph complexity depends on many factors, but the main ones are size, branching, and cyclicity. Some molecular descriptors embrace together all these three parameters, which cannot then be disentangled. The topological index J (and its refinements that include accounting for bond multiplicity and the presence of heteroatoms) was designed to compensate in a significant measure for graph size and cyclicity, and therefore it contains information mainly on branching. In order to separate these factors, two new indices (F and G) related with J are proposed, which allow to group together graphs with the same size into families of constitutional formulas differing in their branching and cyclicity. A comparison with other topological indices revealed that a few other topological indices vary similarly with index G, notably DN2S4 among the triplet indices, and TOTOP among the indices contained in the Molconn-Z program. This comparison involved all possible chemical graphs (i.e. connected planar graphs with vertex degrees not higher than four) with four through six vertices, and all possible alkanes with four through nine carbon atoms.

Prediction of CNS activity of compound libraries using substructure analysis

An in silico ADME/Tox prediction tool based on substructural analysis has been developed. The tool called SUBSTRUCT has been used to predict CNS activity. Data sets with CNS active and nonactive drugs were extracted from the World Drug Index (WDI). The SUBSTRUCT program predicts CNS activity as good as a much more complicated artificial neural network model. SUBSTRUCT separates the data sets with approximately 80% accuracy. Substructural analysis also shows surprisingly large differences in substructure profiles between CNS active and nonactive drugs.

Isothermal Adsorption of Aflatoxin B(1) on HSCAS Clay

Previously, hydrated sodium calcium aluminosilicate (HSCAS), a phyllosilicate clay of the smectite class, was shown to tightly bind aflatoxins and prevent aflatoxicosis in animals. HSCAS is a commercial feed additive that is approved to be used as an anticaking agent. In this study, aflatoxin B(1) (AfB(1)) was isothermally mixed with HSCAS at 15, 25, and 37 degrees C to study the adsorption process. An L adsorption isotherm was characterized from the data, which were fitted to multiple isotherm equations (i.e., Langmuir, multi-Langmuir, general Freundlich, Langmuir-Freundlich combination, Toth, and various transforms of these equations). Information derived from the isotherm equations included capacity, affinity, average capacity, enthalpy of binding, heterogeneity coefficient, multiple site distribution coefficients, and a multisite capacity. Physical characteristics of AfB(1) and HSCAS were measured and modeled to further the understanding of the adsorption process. The data obtained support the hypothesis that AfB(1) chemisorbs to different sites on HSCAS which could include the interlamellar region, edges, and basal surfaces of HSCAS particles.

ALTER: eclectic management of molecular structure data

ALTER is a computer program written to facilitate easy conversion between different representations of molecular structure data. The program functions as a file converter, data generation engine, and through the creation of control or input files, as an interface to other programs. The main aspects of program function--the reading and writing of files; coordinate transformation; data reorganization: structure building; data abstraction, including the generation of a wide variety of topological indices and constitutional descriptors; and display--are described in appropriate detail.

FOLD: integrated analysis and display of protein secondary structure

FOLD, a computer program for the definition and analysis of protein secondary structure, is described. Algorithms implemented in the software are reviewed. These include methods for the identification of simple features such as hydrogen bonds, alpha helices, beta strands, beta bulges, and beta and psi turns. Techniques are also described for the definition and analysis of higher-order structures, such as beta hairpins, beta sheets and their topology, and beta barrels. In addition to considerable textual output the program supports visualization of protein secondary structure in either an atom-based display style or one reproducing the characteristics of a so-called ribbon drawing.

The atom assignment problem in automated de novo drug design. 2. A method for molecular graph and fragment perception

If atom assignment onto 3D molecular graphs is to be optimized, an efficient scheme for placement must be developed. The strategy adopted in this paper is to analyze the molecular graphs in terms of cyclical and non-cyclical nodes; the latter are further divided into terminal and non-terminal nodes. Molecular fragments, from a fragments database, are described in a similar way. A canonical numbering scheme for the fragments and the local subgraph of the molecular graph enables fragments to be placed efficiently onto the molecular graph. Further optimization is achieved by placing similar fragments into bins using a hashing scheme based on the canonical numbering. The graph perception algorithm is illustrated in detail.

A genetic algorithm for the automated generation of molecules within constraints

A genetic algorithm has been designed which generates molecular structures within constraints. The constraints may be any useful function, for example an enzyme active site, a pharmacophore or molecular properties from pattern recognition or rule-induction analyses. The starting point may be random or may utilise known molecules. These are modified to 'grow' into families of structures which, using the evolutionary operators of selection, crossover and mutation evolve to better fit the constraints. The basis of the algorithm is described together with some applications in lead generation, 3D database construction and drug design. Genetic algorithms of this type may have wider applications in chemistry, for example in the design and optimisation of new polymers, materials (e.g. superconducting materials) or synthetic enzymes.

Molecular recognition using a binary genetic search algorithm

A genetic algorithm has been devised and applied to the problems of molecular similarity, pharmacophore elucidation, and determination of molecular conformation. The algorithm is based on a binary representation of molecular position and conformation. Using the genetic operators, crossover, mutation, and selection near optimum conformations and orientations of molecules may be determined which best-fit defined constraints. The constraints may be any useful function for example, intermolecular or intramolecular distances, electrostatic potential on a surface, or volume overlap. Problems with up to 30 degrees of freedom have been tackled successfully.