Inverse QSAR: Reversing Descriptor-Driven Prediction Pipeline Using Attention-Based Conditional Variational Autoencoder.

In order to better foramize it, the notorious inverse-QSAR problem (finding structures of given QSAR-predicted properties) is considered in this paper as a two-step process including (i) finding "seed" descriptor vectors corresponding to user-constrained QSAR model output values and (ii) identifying the chemical structures best matching the "seed" vectors. The main development effort here was focused on the latter stage, proposing a new attention-based conditional variational autoencoder neural-network architecture based on recent developments in attention-based methods. The obtained results show that this workflow was capable of generating compounds predicted to display desired activity while being completely novel compared to the training database (ChEMBL). Moreover, the generated compounds show acceptable druglikeness and synthetic accessibility. Both pharmacophore and docking studies were carried out as "orthogonal" in silico validation methods, proving that some of de novo structures are, beyond being predicted active by 2D-QSAR models, clearly able to match binding 3D pharmacophores and bind the protein pocket.

Chemspace Atlas: Multiscale Chemography of Ultralarge Libraries for Drug Discovery.

Nowadays, drug discovery is inevitably intertwined with the usage of large compound collections. Understanding of their chemotype composition and physicochemical property profiles is of the highest importance for successful hit identification. Efficient polyfunctional tools allowing multifaceted analysis of constantly growing chemical libraries must be Big Data-compatible. Here, we present the freely accessible ChemSpace Atlas (https://chematlas.chimie.unistra.fr), which includes almost 40K hierarchically organized Generative Topographic Maps (GTM) accommodating up to 500 M compounds covering fragment-like, lead-like, drug-like, PPI-like, and NP-like chemical subspaces. They allow users to navigate and analyze ZINC, ChEMBL, and COCONUT from multiple perspectives on different scales: from a bird's eye view of the entire library to structural pattern detection in small clusters. Around 20 physicochemical properties and almost 750 biological activities can be visualized (associated with map zones), supporting activity profiling and analogue search. Moreover, ChemScape Atlas will be extended toward new chemical subspaces (e.g., DNA-encoded libraries and synthons) and functionalities (ADMETox profiling and property-guided de novo compound generation).

HyFactor: A Novel Open-Source, Graph-Based Architecture for Chemical Structure Generation.

Graph-based architectures are becoming increasingly popular as a tool for structure generation. Here, we introduce novel open-source architecture HyFactor in which, similar to the InChI linear notation, the number of hydrogens attached to the heavy atoms was considered instead of the bond types. HyFactor was benchmarked on the ZINC 250K, MOSES, and ChEMBL data sets against conventional graph-based architecture ReFactor, representing our implementation of the reported DEFactor architecture in the literature. On average, HyFactor models contain some 20% less fitting parameters than those of ReFactor. The two architectures display similar validity, uniqueness, and reconstruction rates. Compared to the training set compounds, HyFactor generates more similar structures than ReFactor. This could be explained by the fact that the latter generates many open-chain analogues of cyclic structures in the training set. It has been demonstrated that the reconstruction error of heavy molecules can be significantly reduced using the data augmentation technique. The codes of HyFactor and ReFactor as well as all models obtained in this study are publicly available from our GitHub repository: https://github.com/Laboratoire-de-Chemoinformatique/HyFactor.

HIV-1 drug resistance profiling using amino acid sequence space cartography.

MOTIVATION: Human immunodeficiency virus (HIV) drug resistance is a global healthcare issue. The emergence of drug resistance influenced the efficacy of treatment regimens, thus stressing the importance of treatment adaptation. Computational methods predicting the drug resistance profile from genomic data of HIV isolates are advantageous for monitoring drug resistance in patients. However, existing computational methods for drug resistance prediction are either not suitable for emerging HIV strains with complex mutational patterns or lack interpretability, which is of paramount importance in clinical practice. The approach reported here overcomes these limitations and combines high accuracy of predictions and interpretability of the models. RESULTS: In this work, a new methodology based on generative topographic mapping (GTM) for biological sequence space representation and quantitative genotype-phenotype relationships prediction purposes was introduced. The GTM-based resistance landscapes allowed us to predict the resistance of HIV strains based on sequencing and drug resistance data for three viral proteins [integrase (IN), protease (PR) and reverse transcriptase (RT)] from Stanford HIV drug resistance database. The average balanced accuracy for PR inhibitors was 0.89 ± 0.01, for IN inhibitors 0.85 ± 0.01, for non-nucleoside RT inhibitors 0.73 ± 0.01 and for nucleoside RT inhibitors 0.84 ± 0.01. We have demonstrated in several case studies that GTM-based resistance landscapes are useful for visualization and analysis of sequence space as well as for treatment optimization purposes. Here, GTMs were applied for the in-depth analysis of the relationships between mutation pattern and drug resistance using mutation landscapes. This allowed us to predict retrospectively the importance of the presence of particular mutations (e.g. V32I, L10F and L33F in HIV PR) for the resistance development. This study highlights some perspectives of GTM applications in clinical informatics and particularly in the field of sequence space exploration. AVAILABILITY AND IMPLEMENTATION: https://github.com/karinapikalyova/ISIDASeq. SUPPLEMENTARY INFORMATION: Supplementary data are available at Bioinformatics online.

Atom-to-atom Mapping: A Benchmarking Study of Popular Mapping Algorithms and Consensus Strategies.

In this paper, we compare the most popular Atom-to-Atom Mapping (AAM) tools: ChemAxon,[1] Indigo,[2] RDTool,[3] NameRXN (NextMove),[4] and RXNMapper[5] which implement different AAM algorithms. An open-source RDTool program was optimized, and its modified version ("new RDTool") was considered together with several consensus mapping strategies. The Condensed Graph of Reaction approach was used to calculate chemical distances and develop the "AAM fixer" algorithm for an automatized correction of erroneous mapping. The benchmarking calculations were performed on a Golden dataset containing 1851 manually mapped and curated reactions. The best performing RXNMapper program together with the AMM Fixer was applied to map the USPTO database. The Golden dataset, mapped USPTO and optimized RDTool are available in the GitHub repository https://github.com/Laboratoire-de-Chemoinformatique.

Reaction Data Curation I: Chemical Structures and Transformations Standardization.

The quality of experimental data for chemical reactions is a critical consideration for any reaction-driven study. However, the curation of reaction data has not been extensively discussed in the literature so far. Here, we suggest a 4 steps protocol that includes the curation of individual structures (reactants and products), chemical transformations, reaction conditions and endpoints. Its implementation in Python3 using CGRTools toolkit has been used to clean three popular reaction databases Reaxys, USPTO and Pistachio. The curated USPTO database is available in the GitHub repository (Laboratoire-de-Chemoinformatique/Reaction_Data_Cleaning).

Chemography: Searching for Hidden Treasures.

The days when medicinal chemistry was limited to a few series of compounds of therapeutic interest are long gone. Nowadays, no human may succeed to acquire a complete overview of more than a billion existing or feasible compounds within which the potential "blockbuster drugs" are well hidden and yet only a few mouse clicks away. To reach these "hidden treasures", we adapted the generative topographic mapping method to enable efficient navigation through the chemical space, from a global overview to a structural pattern detection, covering, for the first time, the complete ZINC library of purchasable compounds, relative to 1.6 million biologically relevant ChEMBL molecules. About 40 000 hierarchical maps of the chemical space were constructed. Structural motifs inherent to only one library were identified. Roughly 20 000 off-market ChEMBL compound families represent incentives to enrich commercial catalogs. Alternatively, 125 000 ZINC-specific compound classes, absent in structure-activity bases, are novel paths to explore in medicinal chemistry. The complete list of these chemotypes can be downloaded using the link https://forms.gle/B6bUJj82t9EfmttV6.

Parallel Generative Topographic Mapping: An Efficient Approach for Big Data Handling.

Generative Topographic Mapping (GTM) can be efficiently used to visualize, analyze and model large chemical data. The GTM manifold needs to span the chemical space deemed relevant for a given problem. Therefore, the Frame set (FS) of compounds used for the manifold construction must well cover a given chemical space. Intuitively, the FS size must raise with the size and diversity of the target library. At the same time, the GTM training can be very slow or even becomes technically impossible at FS sizes of the order of 105 compounds - which is a very small number compared to today's commercially accessible compounds, and, especially, to the theoretically feasible molecules. In order to solve this problem, we propose a Parallel GTM algorithm based on the merging of "intermediate" manifolds constructed in parallel for different subsets of molecules. An ensemble of these subsets forms a FS for the "final" manifold. In order to assess the efficiency of the new algorithm, 80 GTMs were built on the FSs of different sizes ranging from 10 to 1.8 M compounds selected from the ChEMBL database. Each GTM was challenged to build classification models for up to 712 biological activities (depending on the FS size). With the novel parallel GTM procedure, we could thus cover the entire spectrum of possible FS sizes, whereas previous studies were forced to rely on the working hypothesis that FS sizes of few thousands of compounds are sufficient to describe the ChEMBL chemical space. In fact, this study formally proves this to be true: a FS containing only 5000 randomly picked compounds is sufficient to represent the entire ChEMBL collection (1.8 M molecules), in the sense that a further increase of FS compound numbers has no benefice impact on the predictive propensity of the above-mentioned 712 activity classification models. Parallel GTM may, however, be required to generate maps based on very large FS, that might improve chemical space cartography of big commercial and virtual libraries, approaching billions of compounds.

Diversifying chemical libraries with generative topographic mapping.

Generative topographic mapping was used to investigate the possibility to diversify the in-house compounds collection of Boehringer Ingelheim (BI). For this purpose, a 2D map covering the relevant chemical space was trained, and the BI compound library was compared to the Aldrich-Market Select (AMS) database of more than 8M purchasable compounds. In order to discover new (sub)structures, the "AutoZoom" tool was developed and applied in order to analyze chemotypes of molecules residing in heavily populated zones of a map and to extract the corresponding maximum common substructures. A set of 401K new structures from the AMS database was retrieved and checked for drug-likeness and biological activity.

Multi-task generative topographic mapping in virtual screening.

The previously reported procedure to generate "universal" Generative Topographic Maps (GTMs) of the drug-like chemical space is in practice a multi-task learning process, in which both operational GTM parameters (example: map grid size) and hyperparameters (key example: the molecular descriptor space to be used) are being chosen by an evolutionary process in order to fit/select "universal" GTM manifolds. After selection (a one-time task aimed at optimizing the compromise in terms of neighborhood behavior compliance, over a large pool of various biological targets), for any further use the manifolds are ready to provide "fit-free" predictive models. Using any structure-activity set-irrespectively whether the associated target served at map fitting stage or not-the generation or "coloring" a property landscape enables predicting the property for any external molecule, with zero additional fitable parameters involved. While previous works have signaled the excellent behavior of such models in aggressive three-fold cross-validation assessments of their predictive power, the present work wished to explore their behavior in Virtual Screening (VS), here simulated on hand of external DUD ligand and decoy series that are fully disjoint from the ChEMBL-extracted landscape coloring sets. Beyond the rather robust results of the universal GTM manifolds in this challenge, it could be shown that the descriptor spaces selected by the evolutionary multi-task learner were intrinsically able to serve as an excellent support for many other VS procedures, starting from parameter-free similarity searching, to local (target-specific) GTM models, to parameter-rich, nonlinear Random Forest and Neural Network approaches.

Mapping of the Available Chemical Space versus the Chemical Universe of Lead-Like Compounds.

This is, to our knowledge, the most comprehensive analysis to date based on generative topographic mapping (GTM) of fragment-like chemical space (40 million molecules with no more than 17 heavy atoms, both from the theoretically enumerated GDB-17 and real-world PubChem/ChEMBL databases). The challenge was to prove that a robust map of fragment-like chemical space can actually be built, in spite of a limited (≪105 ) maximal number of compounds ("frame set") usable for fitting the GTM manifold. An evolutionary map building strategy has been updated with a "coverage check" step, which discards manifolds failing to accommodate compounds outside the frame set. The evolved map has a good propensity to separate actives from inactives for more than 20 external structure-activity sets. It was proven to properly accommodate the entire collection of 40 m compounds. Next, it served as a library comparison tool to highlight biases of real-world molecules (PubChem and ChEMBL) versus the universe of all possible species represented by FDB-17, a fragment-like subset of GDB-17 containing 10 million molecules. Specific patterns, proper to some libraries and absent from others (diversity holes), were highlighted.

Automatized Assessment of Protective Group Reactivity: A Step Toward Big Reaction Data Analysis.

We report a new method to assess protective groups (PGs) reactivity as a function of reaction conditions (catalyst, solvent) using raw reaction data. It is based on an intuitive similarity principle for chemical reactions: similar reactions proceed under similar conditions. Technically, reaction similarity can be assessed using the Condensed Graph of Reaction (CGR) approach representing an ensemble of reactants and products as a single molecular graph, i.e., as a pseudomolecule for which molecular descriptors or fingerprints can be calculated. CGR-based in-house tools were used to process data for 142,111 catalytic hydrogenation reactions extracted from the Reaxys database. Our results reveal some contradictions with famous Greene's Reactivity Charts based on manual expert analysis. Models developed in this study show high accuracy (ca. 90%) for predicting optimal experimental conditions of protective group deprotection.