Inverse mapping of quantum properties to structures for chemical space of small organic molecules

Computer-driven molecular design combines the principles of chemistry, physics, and artificial intelligence to identify chemical compounds with tailored properties. While quantum-mechanical (QM) methods, coupled with machine learning, already offer a direct mapping from 3D molecular structures to their properties, effective methodologies for the inverse mapping in chemical space remain elusive. We address this challenge by demonstrating the possibility of parametrizing a chemical space with a finite set of QM properties. Our proof-of-concept implementation achieves an approximate property-to-structure mapping, the QIM model (which stands for "Quantum Inverse Mapping"), by forcing a variational auto-encoder with a property encoder to obtain a common internal representation for both structures and properties. After validating this mapping for small drug-like molecules, we illustrate its capabilities with an explainability study as well as by the generation of de novo molecular structures with targeted properties and transition pathways between conformational isomers. Our findings thus provide a proof-of-principle demonstration aiming to enable the inverse property-to-structure design in diverse chemical spaces.

GNE-064: A Potent, Selective, and Orally Bioavailable Chemical Probe for the Bromodomains of SMARCA2 and SMARCA4 and the Fifth Bromodomain of PBRM1

Bromodomains are acetyllysine recognition domains present in a variety of human proteins. Bromodomains also bind small molecules that compete with acetyllysine, and therefore bromodomains have been targets for drug discovery efforts. Highly potent and selective ligands with good cellular permeability have been proposed as chemical probes for use in exploring the functions of many of the bromodomain proteins. We report here the discovery of a class of such inhibitors targeting the family VIII bromodomains of SMARCA2 (BRM) and SMARCA4 (BRG1), and PBRM1 (polybromo-1) bromodomain 5. We propose one example from this series, GNE-064, as a chemical probe for the bromodomains SMARCA2, SMARCA4, and PBRM1(5) with the potential for in vivo use.

Nanomolar Synthesis in Droplet Microarrays with UV-Triggered On-Chip Cell Screening

Miniaturization and parallelization of combinatorial organic synthesis is important to accelerate the process of drug discovery while reducing the consumption of reagents and solvents. This work presents a miniaturized platform for on-chip solid-phase combinatorial library synthesis with UV-triggered on-chip cell screening. The platform is based on a nanoporous polymer coating on a glass slide, which is modified via photolithography to yield arrays of hydrophilic (HL) spots surrounded by superhydrophobic (SH) surface. The combination of HL spots and SH background enables confinement of nanoliter droplets, functioning as miniaturized reactors for the solid-phase synthesis. The polymer serves as support for nanomolar solid-phase synthesis, while a photocleavable linker enables the release of the synthesized compounds into the droplets containing live cells. A 588 compound library of bisamides is synthesized via a four-component Ugi reaction on the chip and products are detected via stamping of the droplet array onto a conductive substrate and subsequent matrix-assisted laser desorption ionization mass spectrometry. The light-induced cleavage shows high flexibility in screening conditions by spatial, temporal, and quantitative control.

Cocktailed fragment screening by X-ray crystallography of the antibacterial target undecaprenyl pyrophosphate synthase from Acinetobacter baumannii

Direct soaking of protein crystals with small-molecule fragments grouped into complementary clusters is a useful technique when assessing the potential of a new crystal system to support structure-guided drug discovery. It provides a robustness check prior to any extensive crystal screening, a double check for assay binding cutoffs and structural data for binding pockets that may or may not be picked out in assay measurements. The structural output from this technique for three novel fragment molecules identified to bind to the antibacterial target Acinetobacter baumannii undecaprenyl pyrophosphate synthase are reported, and the different physicochemical requirements of a successful antibiotic are compared with traditional medicines.

chemalot and chemalot_knime: Command line programs as workflow tools for drug discovery

BACKGROUND

Analyzing files containing chemical information is at the core of cheminformatics. Each analysis may require a unique workflow. This paper describes the chemalot and chemalot_knime open source packages. Chemalot is a set of command line programs with a wide range of functionalities for cheminformatics. The chemalot_knime package allows command line programs that read and write SD files from stdin and to stdout to be wrapped into KNIME nodes. The combination of chemalot and chemalot_knime not only facilitates the compilation and maintenance of sequences of command line programs but also allows KNIME workflows to take advantage of the compute power of a LINUX cluster.

RESULTS

Use of the command line programs is demonstrated in three different workflow examples: (1) A workflow to create a data file with project-relevant data for structure-activity or property analysis and other type of investigations, (2) The creation of a quantitative structure-property-relationship model using the command line programs via KNIME nodes, and (3) The analysis of strain energy in small molecule ligand conformations from the Protein Data Bank database.

CONCLUSIONS

The chemalot and chemalot_knime packages provide lightweight and powerful tools for many tasks in cheminformatics. They are easily integrated with other open source and commercial command line tools and can be combined to build new and even more powerful tools. The chemalot_knime package facilitates the generation and maintenance of user-defined command line workflows, taking advantage of the graphical design capabilities in KNIME. Graphical abstract Example KNIME workflow with chemalot nodes and the corresponding command line pipe.

Prospective Assessment of Virtual Screening Heuristics Derived Using a Novel Fusion Score

High-throughput screening (HTS) is a widespread method in early drug discovery for identifying promising chemical matter that modulates a target or phenotype of interest. Because HTS campaigns involve screening millions of compounds, it is often desirable to initiate screening with a subset of the full collection. Subsequently, virtual screening methods prioritize likely active compounds in the remaining collection in an iterative process. With this approach, orthogonal virtual screening methods are often applied, necessitating the prioritization of hits from different approaches. Here, we introduce a novel method of fusing these prioritizations and benchmark it prospectively on 17 screening campaigns using virtual screening methods in three descriptor spaces. We found that the fusion approach retrieves 15% to 65% more active chemical series than any single machine-learning method and that appropriately weighting contributions of similarity and machine-learning scoring techniques can increase enrichment by 1% to 19%. We also use fusion scoring to evaluate the tradeoff between screening more chemical matter initially in lieu of replicate samples to prevent false-positives and find that the former option leads to the retrieval of more active chemical series. These results represent guidelines that can increase the rate of identification of promising active compounds in future iterative screens.

Discovery of a Potent and Selective in Vivo Probe (GNE-272) for the Bromodomains of CBP/EP300

The single bromodomain of the closely related transcriptional regulators CBP/EP300 is a target of much recent interest in cancer and immune system regulation. A co-crystal structure of a ligand-efficient screening hit and the CBP bromodomain guided initial design targeting the LPF shelf, ZA loop, and acetylated lysine binding regions. Structure-activity relationship studies allowed us to identify a more potent analogue. Optimization of permeability and microsomal stability and subsequent improvement of mouse hepatocyte stability afforded 59 (GNE-272, TR-FRET IC₅₀ = 0.02 μM, BRET IC₅₀ = 0.41 μM, BRD4(1) IC₅₀ = 13 μM) that retained the best balance of cell potency, selectivity, and in vivo PK. Compound 59 showed a marked antiproliferative effect in hematologic cancer cell lines and modulates MYC expression in vivo that corresponds with antitumor activity in an AML tumor model.

Plate-based diversity subset screening generation 2: an improved paradigm for high-throughput screening of large compound files

High-throughput screening (HTS) is an effective method for lead and probe discovery that is widely used in industry and academia to identify novel chemical matter and to initiate the drug discovery process. However, HTS can be time consuming and costly and the use of subsets as an efficient alternative to screening entire compound collections has been investigated. Subsets may be selected on the basis of chemical diversity, molecular properties, biological activity diversity or biological target focus. Previously, we described a novel form of subset screening: plate-based diversity subset (PBDS) screening, in which the screening subset is constructed by plate selection (rather than individual compound cherry-picking), using algorithms that select for compound quality and chemical diversity on a plate basis. In this paper, we describe a second-generation approach to the construction of an updated subset: PBDS2, using both plate and individual compound selection, that has an improved coverage of the chemical space of the screening file, whilst only selecting the same number of plates for screening. We describe the validation of PBDS2 and its successful use in hit and lead discovery. PBDS2 screening became the default mode of singleton (one compound per well) HTS for lead discovery in Pfizer.

A Role for Fragment-Based Drug Design in Developing Novel Lead Compounds for Central Nervous System Targets

Hundreds of millions of U.S. dollars are invested in the research and development of a single drug. Lead compound development is an area ripe for new design strategies. Therapeutic lead candidates have been traditionally found using high-throughput in vitro pharmacological screening, a costly method for assaying thousands of compounds. This approach has recently been augmented by virtual screening (VS), which employs computer models of the target protein to narrow the search for possible leads. A variant of VS is fragment-based drug design (FBDD), an emerging in silico lead discovery method that introduces low-molecular weight fragments, rather than intact compounds, into the binding pocket of the receptor model. These fragments serve as starting points for “growing” the lead candidate. Current efforts in virtual FBDD within central nervous system (CNS) targets are reviewed, as is a recent rule-based optimization strategy in which new molecules are generated within a 3D receptor-binding pocket using the fragment as a scaffold. This process not only places special emphasis on creating synthesizable molecules but also exposes computational questions worth addressing. Fragment-based methods provide a viable, relatively low-cost alternative for therapeutic lead discovery and optimization that can be applied to CNS targets to augment current design strategies.

Increasing the delivery of next generation therapeutics from high throughput screening libraries

The pharmaceutical industry has historically relied on high throughput screening as a cornerstone to identify chemical equity for drug discovery projects. However, with pharmaceutical companies moving through a phase of diminished returns and alternative hit identification strategies proving successful, it is more important than ever to understand how this approach can be used more effectively to increase the delivery of next generation therapeutics from high throughput screening libraries. There is a wide literature that describes HTS and fragment based screening approaches which offer clear direction on the process for these two distinct activities. However, few people have considered how best to identify medium to low molecular weight compounds from large diversity screening sets and increase downstream success.

Atom-Atom-Path similarity and Sphere Exclusion clustering: tools for prioritizing fragment hits

Background

After performing a fragment based screen the resulting hits need to be prioritized for follow-up structure elucidation and chemistry. This paper describes a new similarity metric, Atom-Atom-Path (AAP) similarity that is used in conjunction with the Directed Sphere Exclusion (DISE) clustering method to effectively organize and prioritize the fragment hits. The AAP similarity rewards common substructures and recognizes minimal structure differences. The DISE method is order-dependent and can be used to enrich fragments with properties of interest in the first clusters.

Results

The merit of the software is demonstrated by its application to the MAP4K4 fragment screening hits using ligand efficiency (LE) as quality measure. The first clusters contain the hits with the highest LE. The clustering results can be easily visualized in a LE-over-clusters scatterplot with points colored by the members’ similarity to the corresponding cluster seed. The scatterplot enables the extraction of preliminary SAR.

Conclusions

The detailed structure differentiation of the AAP similarity metric is ideal for fragment-sized molecules. The order-dependent nature of the DISE clustering method results in clusters ordered by a property of interest to the teams. The combination of both allows for efficient prioritization of fragment hit for follow-ups.

Graphical abstract