Exploring the GDB-13 chemical space using deep generative models

Recent applications of recurrent neural networks (RNN) enable training models that sample the chemical space. In this study we train RNN with molecular string representations (SMILES) with a subset of the enumerated database GDB-13 (975 million molecules). We show that a model trained with 1 million structures (0.1% of the database) reproduces 68.9% of the entire database after training, when sampling 2 billion molecules. We also developed a method to assess the quality of the training process using negative log-likelihood plots. Furthermore, we use a mathematical model based on the “coupon collector problem” that compares the trained model to an upper bound and thus we are able to quantify how much it has learned. We also suggest that this method can be used as a tool to benchmark the learning capabilities of any molecular generative model architecture. Additionally, an analysis of the generated chemical space was performed, which shows that, mostly due to the syntax of SMILES, complex molecules with many rings and heteroatoms are more difficult to sample.

PubChem and ChEMBL beyond Lipinski

Seven million of the currently 94 million entries in the PubChem database break at least one of the four Lipinski constraints for oral bioavailability, 183,185 of which are also found in the ChEMBL database. These non-Lipinski PubChem (NLP) and ChEMBL (NLC) subsets are interesting because they contain new modalities that can display biological properties not accessible to small molecule drugs. Unfortunately, the current search tools in PubChem and ChEMBL are designed for small molecules and are not well suited to explore these subsets, which therefore remain poorly appreciated. Herein we report MXFP (macromolecule extended atom-pair fingerprint), a 217-D fingerprint tailored to analyze large molecules in terms of molecular shape and pharmacophores. We implement MXFP in two web-based applications, the first one to visualize NLP and NLC interactively using Faerun (http://faerun.gdb.tools/), the second one to perform MXFP nearest neighbor searches in NLP and NLC (http://similaritysearch.gdb.tools/). We show that these tools provide a meaningful insight into the diversity of large molecules in NLP and NLC. The interactive tools presented here are publicly available at http://gdb.unibe.ch and can be used freely to explore and better understand the diversity of non-Lipinski molecules in PubChem and ChEMBL.

FUn: a framework for interactive visualizations of large, high-dimensional datasets on the web

Motivation

During the past decade, big data have become a major tool in scientific endeavors. Although statistical methods and algorithms are well-suited for analyzing and summarizing enormous amounts of data, the results do not allow for a visual inspection of the entire data. Current scientific software, including R packages and Python libraries such as ggplot2, matplotlib and plot.ly, do not support interactive visualizations of datasets exceeding 100 000 data points on the web. Other solutions enable the web-based visualization of big data only through data reduction or statistical representations. However, recent hardware developments, especially advancements in graphical processing units, allow for the rendering of millions of data points on a wide range of consumer hardware such as laptops, tablets and mobile phones. Similar to the challenges and opportunities brought to virtually every scientific field by big data, both the visualization of and interaction with copious amounts of data are both demanding and hold great promise.

Results

Here we present FUn, a framework consisting of a client (Faerun) and server (Underdark) module, facilitating the creation of web-based, interactive 3D visualizations of large datasets, enabling record level visual inspection. We also introduce a reference implementation providing access to SureChEMBL, a database containing patent information on more than 17 million chemical compounds.

Availability and implementation

The source code and the most recent builds of Faerun and Underdark, Lore.js and the data preprocessing toolchain used in the reference implementation, are available on the project website (http://doc.gdb.tools/fun/).

Contact

daniel.probst@dcb.unibe.ch or jean-louis.reymond@dcb.unibe.ch.

Web-Based Tools for Polypharmacology Prediction

Drug promiscuity or polypharmacology is the ability of small molecules to interact with multiple protein targets simultaneously. In drug discovery, understanding the polypharmacology of potential drug molecules is crucial to improve their efficacy and safety, and to discover the new therapeutic potentials of existing drugs. Over the past decade, several computational methods have been developed to study the polypharmacology of small molecules, many of which are available as Web services. In this chapter, we review some of these Web tools focusing on ligand based approaches. We highlight in particular our recently developed polypharmacology browser (PPB) and its application for finding the side targets of a new inhibitor of the TRPV6 calcium channel.

Polypharmacology Browser PPB2: Target Prediction Combining Nearest Neighbors with Machine Learning

Here we report PPB2 as a target prediction tool assigning targets to a query molecule based on ChEMBL data. PPB2 computes ligand similarities using molecular fingerprints encoding composition (MQN), molecular shape and pharmacophores (Xfp), and substructures (ECfp4) and features an unprecedented combination of nearest neighbor (NN) searches and Naı̈ve Bayes (NB) machine learning, together with simple NN searches, NB and Deep Neural Network (DNN) machine learning models as further options. Although NN(ECfp4) gives the best results in terms of recall in a 10-fold cross-validation study, combining NN searches with NB machine learning provides superior precision statistics, as well as better results in a case study predicting off-targets of a recently reported TRPV6 calcium channel inhibitor, illustrating the value of this combined approach. PPB2 is available to assess possible off-targets of small molecule drug-like compounds by public access at http://gdb.unibe.ch .

A probabilistic molecular fingerprint for big data settings

Background

Among the various molecular fingerprints available to describe small organic molecules, extended connectivity fingerprint, up to four bonds (ECFP4) performs best in benchmarking drug analog recovery studies as it encodes substructures with a high level of detail. Unfortunately, ECFP4 requires high dimensional representations (≥ 1024D) to perform well, resulting in ECFP4 nearest neighbor searches in very large databases such as GDB, PubChem or ZINC to perform very slowly due to the curse of dimensionality.

Results

Herein we report a new fingerprint, called MinHash fingerprint, up to six bonds (MHFP6), which encodes detailed substructures using the extended connectivity principle of ECFP in a fundamentally different manner, increasing the performance of exact nearest neighbor searches in benchmarking studies and enabling the application of locality sensitive hashing (LSH) approximate nearest neighbor search algorithms. To describe a molecule, MHFP6 extracts the SMILES of all circular substructures around each atom up to a diameter of six bonds and applies the MinHash method to the resulting set. MHFP6 outperforms ECFP4 in benchmarking analog recovery studies. By leveraging locality sensitive hashing, LSH approximate nearest neighbor search methods perform as well on unfolded MHFP6 as comparable methods do on folded ECFP4 fingerprints in terms of speed and relative recovery rate, while operating in very sparse and high-dimensional binary chemical space.

Conclusion

MHFP6 is a new molecular fingerprint, encoding circular substructures, which outperforms ECFP4 for analog searches while allowing the direct application of locality sensitive hashing algorithms. It should be well suited for the analysis of large databases. The source code for MHFP6 is available on GitHub (https://github.com/reymond-group/mhfp).

Electronic supplementary material

The online version of this article (10.1186/s13321-018-0321-8) contains supplementary material, which is available to authorized users.

Identifying Lysophosphatidic Acid Acyltransferase β (LPAAT-β) as the Target of a Nanomolar Angiogenesis Inhibitor from a Phenotypic Screen Using the Polypharmacology Browser PPB2

By screening a focused library of kinase inhibitor analogues in a phenotypic co-culture assay for angiogenesis inhibition, we identified an aminotriazine that acts as a cytostatic nanomolar inhibitor. However, this aminotriazine was found to be completely inactive in a whole-kinome profiling assay. To decipher its mechanism of action, we used the online target prediction tool PPB2 (http://ppb2.gdb.tools), which suggested lysophosphatidic acid acyltransferase β (LPAAT-β) as a possible target for this aminotriazine as well as several analogues identified by structure-activity relationship profiling. LPAAT-β inhibition (IC₅₀ ≈15 nm) was confirmed in a biochemical assay and by its effects on cell proliferation in comparison with a known LPAAT-β inhibitor. These experiments illustrate the value of target-prediction tools to guide target identification for phenotypic screening hits and significantly expand the rather limited pharmacology of LPAAT-β inhibitors.

Structure of bacterial oligosaccharyltransferase PglB bound to a reactive LLO and an inhibitory peptide

Oligosaccharyltransferase (OST) is a key enzyme of the N-glycosylation pathway, where it catalyzes the transfer of a glycan from a lipid-linked oligosaccharide (LLO) to an acceptor asparagine within the conserved sequon N-X-T/S. A previous structure of a ternary complex of bacterial single subunit OST, PglB, bound to a non-hydrolyzable LLO analog and a wild type acceptor peptide showed how both substrates bind and how an external loop (EL5) of the enzyme provided specific substrate-binding contacts. However, there was a relatively large separation of the substrates at the active site. Here we present the X-ray structure of PglB bound to a reactive LLO analog and an inhibitory peptide, revealing previously unobserved interactions in the active site. We found that the atoms forming the N-glycosidic bond (C-1 of the GlcNAc moiety of LLO and the –NH₂ group of the peptide) are closer than in the previous structure, suggesting that we have captured a conformation closer to the transition state of the reaction. We find that the distance between the divalent metal ion and the glycosidic oxygen of LLO is now 4 Å, suggesting that the metal stabilizes the leaving group of the nucleophilic substitution reaction. Further, the carboxylate group of a conserved aspartate of PglB mediates an interaction network between the reducing-end sugar of the LLO, the asparagine side chain of the acceptor peptide, and a bound divalent metal ion. The interactions identified in this novel state are likely to be relevant in the catalytic mechanisms of all OSTs.

Novel peptide-dendrimer/lipid/oligonucleotide ternary complexes for efficient cellular uptake and improved splice-switching activity

Despite the advances in gene therapy and in oligonucleotide (ON) chemistry, efficient cellular delivery remains an obstacle. Most current transfection reagents suffer from low efficacy or high cytotoxicity. In this report, we describe the synergism between lipid and dendrimer delivery vectors to enhance the transfection efficiency, while avoiding high toxicity. We screened a library of 20 peptide dendrimers representing three different generations and evaluated their capability to deliver a single-stranded splice-switching ON after formulating with lipids (DOTMA/DOPE). The transfection efficiency was analyzed in 5 reporter cell lines, in serum-free and serum conditions, and with 5 different formulation protocols. All formulations displayed low cytotoxicity to the majority of the tested cell lines. The complex sizes were < 200 nm; particle size distributions of effective mixtures were < 80 nm; and, the zeta potential was dependent on the formulation buffer used. The best dendrimer enhanced transfection in a HeLa reporter cell line by 30-fold compared to untreated cells under serum-free conditions. Interestingly, addition of sucrose to the formulation enabled - for the first time - peptide dendrimers/lipid complexes to efficiently deliver splice-switching ON in the presence of serum, reaching 40-fold increase in splice switching. Finally, in vivo studies highlighted the potential of these formulae to change the biodistribution pattern to be more towards the liver (90% of injected dose) compared to the kidneys (5% of injected dose) or to unformulated ON. This success encourages further development of peptide dendrimer complexes active in serum and future investigation of mechanisms behind the influence of additives on transfection efficacy.

Exploring DrugBank in Virtual Reality Chemical Space

The recent general availability of low-cost virtual reality headsets and accompanying three-dimensional (3D) engine support presents an opportunity to bring the concept of chemical space into virtual environments. While virtual reality applications represent a category of widespread tools in other fields, their use in the visualization and exploration of abstract data such as chemical spaces has been experimental. In our previous work, we established the concept of interactive two-dimensional (2D) maps of chemical spaces followed by interactive web-based 3D visualizations, culminating in the interactive web-based 3D visualization of extremely large chemical spaces. Virtual reality chemical spaces are a natural extension of these concepts. As 2D and 3D embeddings and projections of high-dimensional chemical fingerprint spaces have been shown to be valuable tools in chemical space visualization and exploration, existing pipelines of data mining and preparation can be extended to be used in virtual reality applications. Here we present an application based on the Unity engine and the Virtual Reality Toolkit, allowing for the interactive exploration of chemical space populated by DrugBank compounds in virtual reality. The source code of the application as well as the most recent build are available on GitHub ( https://github.com/reymond-group/virtual-reality-chemical-space ).

Synthesis of Lipid-Linked Oligosaccharides (LLOs) and Their Phosphonate Analogues as Probes To Study Protein Glycosylation Enzymes

Here we review chemical and chemoenzymatic methods for the synthesis of lipid-linked oligosaccharides (LLOs) and their phosphonate analogues, which serve as substrates and inhibitors to investigate the structure and mechanism of protein N-glycosylation enzymes. We emphasize how to overcome the challenges pertaining to the instability and difficult physicochemical properties of this class of compounds.

1 Introduction

2 LLO Syntheses

2.1 Glycosyl Phosphate Syntheses

2.2 Glycosyl Phosphonates

2.3 Lipid Elongation

2.4 Lipid Phosphates

2.5 Coupling Reaction Strategies

3 Chemoenzymatic Synthesis of Elongated LLOs

4 Biological Properties of Synthetic LLOs

5 Conclusion

Optimizing Antimicrobial Peptide Dendrimers in Chemical Space

We used nearest-neighbor searches in chemical space to improve the activity of the antimicrobial peptide dendrimer (AMPD) G3KL and identified dendrimer T7, which has an expanded activity range against Gram-negative pathogenic bacteria including Klebsiellae pneumoniae, increased serum stability, and promising activity in an in vivo infection model against a multidrug-resistant strain of Acinetobacter baumannii. Imaging, spectroscopic studies, and a structural model from molecular dynamics simulations suggest that T7 acts through membrane disruption. These experiments provide the first example of using virtual screening in the field of dendrimers and show that dendrimer size does not limit the activity of AMPDs.

An antimicrobial bicyclic peptide from chemical space against multidrug resistant Gram-negative bacteria

We used the concept of chemical space to explore a virtual library of bicyclic peptides formed by double thioether cyclization of a precursor linear peptide, and identified an antimicrobial bicyclic peptide (AMBP) with remarkable activity against several MDR strains of Acinetobacter baumannii and Pseudomonas aeruginosa.

Characterization of the single-subunit oligosaccharyltransferase STT3A from Trypanosoma brucei using synthetic peptides and lipid-linked oligosaccharide analogs

The initial transfer of a complex glycan in protein N-glycosylation is catalyzed by oligosaccharyltransferase (OST), which is generally a multisubunit membrane protein complex in the endoplasmic reticulum but a single-subunit enzyme (ssOST) in some protists. To investigate the reaction mechanism of ssOST, we recombinantly expressed, purified and characterized the STT3A protein from Trypanosoma brucei (TbSTT3A). We analyzed the in vitro activity of TbSTT3A by synthesizing fluorescently labeled acceptor peptides as well as lipid-linked oligosaccharide (LLO) analogs containing a chitobiose moiety coupled to oligoprenyl carriers of distinct lengths (C₁₀, C₁₅, C₂₀ and C₂₅) and with different double bond stereochemistry. We found that in addition to proline, charged residues at the +1 position of the sequon inhibited glycan transfer. An acidic residue at the −2 position significantly increased catalytic turnover but was not essential, in contrast to the bacterial OST. While all synthetic LLO analogs were processed by TbSTT3A, the length of the polyprenyl tail, but not the stereochemistry of the double bonds, determined their apparent affinity. We also synthesized phosphonate analogs of the LLOs, which were found to be competitive inhibitors of the reaction, although with lower apparent affinity to TbSTT3A than the active pyrophosphate analogs.

Structural basis of the molecular ruler mechanism of a bacterial glycosyltransferase

The membrane-associated, processive and retaining glycosyltransferase PglH from Campylobacter jejuni is part of the biosynthetic pathway of the lipid-linked oligosaccharide (LLO) that serves as the glycan donor in bacterial protein N-glycosylation. Using an unknown counting mechanism, PglH catalyzes the transfer of exactly three α1,4 N-acetylgalactosamine (GalNAc) units to the growing LLO precursor, GalNAc-α1,4-GalNAc-α1,3-Bac-α1-PP-undecaprenyl. Here, we present crystal structures of PglH in three distinct states, including a binary complex with UDP-GalNAc and two ternary complexes containing a chemo-enzymatically generated LLO analog and either UDP or synthetic, nonhydrolyzable UDP-CH₂-GalNAc. PglH contains an amphipathic helix ("ruler helix") that has a dual role of facilitating membrane attachment and glycan counting. The ruler helix contains three positively charged side chains that can bind the pyrophosphate group of the LLO substrate and thus limit the addition of GalNAc units to three. These results, combined with molecular dynamics simulations, provide the mechanism of glycan counting by PglH.

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.

SmilesDrawer: Parsing and Drawing SMILES-Encoded Molecular Structures Using Client-Side JavaScript

Here we present SmilesDrawer, a dependency-free JavaScript component capable of both parsing and drawing SMILES-encoded molecular structures client-side, developed to be easily integrated into web projects and to display organic molecules in large numbers and fast succession. SmilesDrawer can draw structurally and stereochemically complex structures such as maitotoxin and C₆₀ without using templates, yet has an exceptionally small computational footprint and low memory usage without the requirement for loading images or any other form of client-server communication, making it easy to integrate even in secure (intranet, firewalled) or offline applications. These features allow the rendering of thousands of molecular structure drawings on a single web page within seconds on a wide range of hardware supporting modern browsers. The source code as well as the most recent build of SmilesDrawer is available on Github ( http://doc.gdb.tools/smilesDrawer/ ). Both yarn and npm packages are also available.

Lipidated Peptide Dendrimers Killing Multidrug-Resistant Bacteria

New antibiotics are urgently needed to address multidrug-resistant (MDR) bacteria. Herein we report that second-generation (G2) peptide dendrimers bearing a fatty acid chain at the dendrimer core efficiently kill Gram-negative bacteria including Pseudomonas aeruginosa and Acinetobacter baumannii, two of the most problematic MDR bacteria worldwide. Our most active dendrimer TNS18 is also active against Gram-positive methicillin-resistant Staphylococcus aureus. Based on circular dichroism and molecular dynamics studies, we hypothesize that TNS18 adopts a hydrophobically collapsed conformation in water with the fatty acid chain backfolded onto the peptide dendrimer branches and that the dendrimer unfolds in contact with the membrane to expose its lipid chain and hydrophobic residues, thereby facilitating membrane disruption leading to rapid bacterial cell death. Dendrimer TNS18 shows promising in vivo activity against MDR clinical isolates of A. baumannii and Escherichia coli, suggesting that lipidated peptide dendrimers might become a new class of antibacterial agents.

Virtual Exploration of the Ring Systems Chemical Universe

Here, we explore the chemical space of all virtually possible organic molecules focusing on ring systems, which represent the cyclic cores of organic molecules obtained by removing all acyclic bonds and converting all remaining atoms to carbon. This approach circumvents the combinatorial explosion encountered when enumerating the molecules themselves. We report the chemical universe database GDB4c containing 916 130 ring systems up to four saturated or aromatic rings and maximum ring size of 14 atoms and GDB4c3D containing the corresponding 6 555 929 stereoisomers. Almost all (98.6%) of these ring systems are unknown and represent chiral 3D-shaped macrocycles containing small rings and quaternary centers reminiscent of polycyclic natural products. We envision that GDB4c can serve to select new ring systems from which to design analogs of such natural products. The database is available for download at www.gdb.unibe.ch together with interactive visualization and search tools as a resource for molecular design.

Molecular basis of lipid-linked oligosaccharide recognition and processing by bacterial oligosaccharyltransferase

Oligosaccharyltransferase (OST) is a membrane-integral enzyme that catalyzes the transfer of glycans from lipid-linked oligosaccharides (LLOs) onto asparagine side chains, the first step in protein N-glycosylation. Here, we report the X-ray structure of a single-subunit OST, PglB from Campylobacter lari, trapped in an intermediate state bound to an acceptor peptide and a synthetic LLO analog. The structure reveals the role of the external loop EL5, present in all OST enzymes, in substrate recognition. Whereas the N-terminal half of EL5 binds LLO, the C-terminal half interacts with the acceptor peptide. The glycan moiety of LLO must thread under EL5 to access the active site. Reducing EL5 mobility decreases the catalytic rate of OST when full-size heptasaccharide LLO is provided, but not for a monosaccharide-containing LLO analog. Our results define the chemistry of a ternary complex state, assign functional roles to conserved OST motifs, and provide opportunities for glycoengineering by rational design of PglB.