High-throughput Screening Campaign Research Articles

Optimization of Selectivity and Pharmacokinetic Properties of Salt-Inducible Kinase Inhibitors that Led to the Discovery of Pan-SIK Inhibitor GLPG3312.

Salt-inducible kinases (SIKs) SIK1, SIK2, and SIK3 are serine/threonine kinases and form a subfamily of the protein kinase AMP-activated protein kinase (AMPK) family. Inhibition of SIKs in stimulated innate immune cells and mouse models has been associated with a dual mechanism of action consisting of a reduction of pro-inflammatory cytokines and an increase of immunoregulatory cytokine production, suggesting a therapeutic potential for inflammatory diseases. Following a high-throughput screening campaign, subsequent hit to lead optimization through synthesis, structure-activity relationship, kinome selectivity, and pharmacokinetic investigations led to the discovery of clinical candidate GLPG3312 (compound 28), a potent and selective pan-SIK inhibitor (IC50: 2.0 nM for SIK1, 0.7 nM for SIK2, and 0.6 nM for SIK3). Characterization of the first human SIK3 crystal structure provided an understanding of the binding mode and kinome selectivity of the chemical series. GLPG3312 demonstrated both anti-inflammatory and immunoregulatory activities in vitro in human primary myeloid cells and in vivo in mouse models.

Experimental structure based drug design (SBDD) applications for anti-leishmanial drugs: A paradigm shift?

Leishmaniasis is a group of neglected tropical diseases caused by at least 20 species of Leishmania protozoa, which are spread by the bite of infected sandflies. There are three main forms of the disease: cutaneous leishmaniasis (CL, the most common), visceral leishmaniasis (VL, also known as kala-azar, the most serious), and mucocutaneous leishmaniasis. One billion people live in areas endemic to leishmaniasis, with an annual estimation of 30,000 new cases of VL and more than 1 million of CL. New treatments for leishmaniasis are an urgent need, as the existing ones are inefficient, toxic, and/or expensive. We have revised the experimental structure-based drug design (SBDD) efforts applied to the discovery of new drugs against leishmaniasis. We have grouped the explored targets according to the metabolic pathways they belong to, and the key achieved advances are highlighted and evaluated. In most cases, SBDD studies follow high-throughput screening campaigns and are secondary to pharmacokinetic optimization, due to the majoritarian belief that there are few validated targets for SBDD in leishmaniasis. However, some SBDD strategies have significantly contributed to new drug candidates against leishmaniasis and a bigger number holds promise for future development.

Abstract IA013: Integrating natural product samples back into drug screening programs

Abstract Natural products have a proven track record in use as anti-cancer agents. As of 2020 natural products (NPs), NP pharmacophores, and NP mimics represent a significant percentage of small molecule-based anti-cancer drugs. However, accessing the potential of NPs in high-throughput screening (HTS) campaigns has been challenging due to the complexity of the crude NP extracts, the time necessary for follow-up isolation efforts, and indefinite capabilities for resupply. This has resulted in NP-based libraries being significantly underrepresented in most recent large-scale HTS programs. The US National Cancer Institute’s Natural Product Extract Repository is one of the world’s largest, most diverse collections of natural products containing >200,000 unique extracts derived from plant, marine and microbial organisms. To address the challenges in screening NPs and encourage increased assay of NP chemical diversity, the NCI created a Program for Natural Product Discovery (NPNPD) which has developed an automated, high-throughput robotics platform that has created 500,000 pre-fractionated NP samples and released them to the public for screening. In addition, automated secondary purification and structure elucidation systems were created that significantly decrease costs and timelines. This presentation will discuss the reasons for, challenges to, and mechanisms by which the NCI is working to increase the pace and efficiency of NP-based drug discovery. Citation Format: Barry R. O'Keefe. Integrating natural product samples back into drug screening programs [abstract]. In: Proceedings of the AACR-NCI-EORTC Virtual International Conference on Molecular Targets and Cancer Therapeutics; 2023 Oct 11-15; Boston, MA. Philadelphia (PA): AACR; Mol Cancer Ther 2023;22(12 Suppl):Abstract nr IA013.

Discovery of a Novel Benzodiazepine Series of Cbl-b Inhibitors for the Enhancement of Antitumor Immunity.

Casitas B-lineage lymphoma proto-oncogene-b (Cbl-b) is a RING finger E3 ligase that is responsible for repressing T-cell, natural killer (NK) cell, and B-cell activation. The robust antitumor activity observed in Cbl-b deficient mice arising from elevated T-cell and NK-cell activity justified our discovery effort toward Cbl-b inhibitors that might show therapeutic promise in immuno-oncology, where activation of the immune system can drive the recognition and killing of cancer cells. We undertook a high-throughput screening campaign followed by structure-enabled optimization to develop a novel benzodiazepine series of potent Cbl-b inhibitors. This series displayed nanomolar levels of biochemical potency, as well as potent T-cell activation. The functional activity of this class of Cbl-b inhibitors was further corroborated with ubiquitin-based cellular assays.

A high-throughput screening strategy identifies adenosine 5'-monophosphate as a novel enhancing molecule of the cardio-protective enzyme 5-oxoprolinase (OPLAH)

Abstract Background/Introduction Oxidative stress plays an essential role in the pathophysiology of heart failure (HF), and reducing it might improve outcomes in patients suffering from it. We recently identified 5-oxoproline to be elevated in patients with HF and that the increased levels are associated with a poor prognosis. 5-oxoprolinase (OPLAH) is an enzyme that converts toxic 5-oxoproline to harmless glutamate. We previously showed that the overexpression of OPLAH prevented cardiac damage in experimental models. Purpose To identify small molecules that hold the potential to increase the activity of the antioxidant enzyme OPLAH. Methods and results OPLAH expression vectors were established containing a 6xHis-tag in the N-terminal region and were used to efficiently overexpress enzymatically active recombinant OPLAH protein (rOPLAH) in human embryonic kidney cells (HEK293A) and Spodoptera frugiperda cells (SF9). rOPLAH expressing His-tag was purified under native conditions using an affinity chromatography matrix, and enzymatic activity was proven by measuring the conversion of 5-oxoproline to glutamate using LC-MS/MS. After miniaturization and optimization of a luminescence-based technique that measures OPLAH's ATPase activity, a high-throughput screening campaign of 1280 FDA-approved compounds was performed. Adenosine 5′-monophosphate (5'-AMP) was identified to enhance the ATPase activity of OPLAH by >50%. In addition, an LC-MS/MS dose-response curve exhibited that increasing 5'-AMP concentration resulted in a higher glutamate production using isolated rOPLAH. A thermal shift assay confirmed the interaction of 5'-AMP and OPLAH. Since 5’-AMP is unable to cross the cellular plasma membrane, acetalated dextran spermine-modified nanoparticles loaded with 5'-AMP were produced (AcDXSp@AMP NPs) to target cells. NPs were prepared by the water-in-oil-in-water double emulsion technique. Particle size, polydispersity index and z–potential of the NPs were characterized by dynamic and electrophoretic light scattering. HPLC tested the quantification of encapsulated 5'-AMP and its release over time. The efficacy of OPLAH activation by 5'-AMP-loaded NPs was tested in HeLa cells cultured in the presence of 10 mM 13C-5-oxoproline. Employing AcDXSp@AMP NPs for 12h to cells resulted in about 50% elevation of 13C-glutamate production, showing successful activation of OPLAH by AMP in cells. Conclusion Here, we developed a high-throughput screening for 5-oxprolinase and identified a non-cell permeable enhancer compound, adenosine 5′-monophosphate. Orthogonal studies confirmed 5'-AMP affinity to OPLAH. Finally, we successfully utilized nanoparticles as a drug-delivery system for 5'-AMP and proved its intracellular effect. Our efforts are a blueprint for bringing target discovery to treatment identification. As such, nanoparticle-based delivery systems hold great potential for future trials utilizing impermeable modulators to target heart failure and other diseases.Illustrative workflow

Discovery of First-in-Class Small Molecule Inhibitors of the IRF4-PU.1/Spi-B Interaction

Introduction Many transcription factors (TF) are promising therapeutic targets in hematological malignancies but have been mislabeled as ‘undruggable’ due to their dynamic structures and lack of feasible paths for bioactive small molecules. Multiple myeloma (MM), the second most common hematological malignancy, is predominantly driven by oncogenic transcription factors such as c-Myc & interferon regulatory factor 4 (IRF4). Attempts at targeting IRF4 have relied on immunomodulatory drugs (IMiDs), such as lenalidomide and pomalidomide, which target upstream TFs of IRF4 - IKZF1 & IKZF3. Unfortunately, IMiDs are highly susceptible to resistance mechanisms in the clinic and fail to provide deep and sustained downregulation of IRF4 due to theirindirect mechanism of action. Currently, no known small molecules that directly bind and inhibit IRF4 exist. Moreover, no probes or screening based assays have been developed for IRF4 or any member of the IRF family.IRF4 is the top ranked dependency of MM and is a validated clinical target given the success of IMiD-based regimens. Thus, orally available small molecules that can directly target IRF4 will have significant benefit for patients with MM. Methods & Results To identify small molecules that inhibit or bind IRF4, we first cloned, expressed, and purified numerous constructs of the interferon accessory domain (IAD - residues 238-420) and full-length IRF4 (residues 1-451). To enable traditional high-throughput screening (HTS) campaigns, we first identified peptide ligands of IRF4 based on the IRF4 - PU.1/Spi-B interaction, an oncogenic signaling node in various hematological malignancies. We profiled peptides derived from PU.1 and Spi-B and confirmed binding to the IAD domain with K ds of 1 - 7 µM by isothermal titration calorimetry (ITC) and fluorescence polarization (FP). Encouraged by the robust signal seen with our FP assay, we further developed a TR-FRET based method using a CoraFluor labeled IAD domain which significantly boosted our signal-to-noise ratio for our HTS efforts (Z' > 0.8). We screened this interaction against the Chembridge2020 drug library, which consists of 50,000 lead-like small molecules. The screen afforded 68 small molecules that had inhibition > 50% by TR-FRET (hit rate of ~0.1%). We prioritized our most potent compound, HIT-1, which has a TR-FRET IC 50 of 2 µM. Further binding validation including ITC, biotin-based pull-downs in human myeloma cell lines (HMCLs), and a BRET based target engagement assay confirm binding to both recombinant and full-length IRF4. Additionally, medicinal chemistry optimization led to the development of 1, with sub-micromolar affinity to IRF4, and an inactive control 2. Biological assessment of compounds 1 and inactive control 2 are underway in HMCLs; initial results show on-target modulation of IRF4 signaling processes. Compound 1 represents the first validated small molecule targeting the ‘undruggable’ IRF family and warrants its use in investigating the IRF4-PU.1/Spi-B interaction in oncology and immunology. Conclusions These small molecules are the first inhibitors of the IRF4-PU.1 interaction. We have developed novel biochemical assays that can be applied to other IRF family members, including IRF8. Additionally, we showcase translatable strategies to build novel biochemical assays for other transcription factors, leveraging known PPIs for probe development. These ligands represent starting points for hetero-bifunctional compounds and chemical probes for modulating IRF4 activity both in vitro and in vivo.

High-Throughput cell-based immunofluorescence assays against influenza

A rapid drug discovery response to influenza outbreaks with the potential to reach pandemic status could help minimize the virus's impact by reducing the time to identify anti-influenza drugs. Although several anti-influenza strategies have been considered in the search for new drugs, only a few therapeutic agents are approved for clinical use. The cytopathic effect induced by the influenza virus in Madin Darby canine kidney (MDCK) cells has been widely used for high-throughput anti-influenza drug screening, but the fact that the MDCK cells are not human cells constitutes a disadvantage when searching for new therapeutic agents for human use. We have developed a highly sensitive cell-based imaging assay for the identification of inhibitors of influenza A and B virus that is high-throughput compatible using the A549 human cell line. The assay has also been optimized for the assessment of the neutralizing effect of anti-influenza antibodies in the absence of trypsin, which allows testing of purified antibodies and serum samples. This assay platform can be applied to full high-throughput screening campaigns or later stages requiring quantitative potency determinations for structure-activity relationships.

Fused Tetrahydroquinolines Are Interfering with Your Assay.

Tricyclic tetrahydroquinolines (THQs) have been repeatedly reported as hits across a diverse range of high-throughput screening (HTS) campaigns. The activities of these compounds, however, are likely due to reactive byproducts that interfere with the assay. As a lesser studied class of pan-assay interference compounds, the mechanism by which fused THQs react with protein targets remains largely unknown. During HTS follow-up, we characterized the behavior and stability of several fused tricyclic THQs. We synthesized key analogues to pinpoint the cyclopentene ring double bond as a source of reactivity of fused THQs. We found that these compounds degrade in solution under standard laboratory conditions in days. Importantly, these observations make it likely that fused THQs, which are ubiquitously found within small molecule screening libraries, are unlikely the intact parent compounds. We urge deprioritization of tricylic THQ hits in HTS follow-up and caution against the investment of resources to follow-up on these problematic compounds.

Identification of novel glucocerebrosidase chaperones by unexpected skeletal rearrangement reaction

Compound 5 was identified from a high-throughput screening campaign as a small molecule pharmacological chaperone of glucocerebrocidase (GCase), a lysosomal hydrolase encoded by the GBA1 gene, variants of which are associated with Gaucher disease and Parkinson’s disease. Further investigations revealed that compound 5 was slowly transformed into a regio-isomeric compound (6) in PBS buffer, plausibly via a ring-opening at hemiaminal moiety accompanied by subsequent intramolecular CC bond formation. Utilising this unexpected skeletal rearrangement reaction, a series of compound 6 analogues was synthesized which yielded multiple potent GCase pharmacological chaperones with sub-micromolar EC50 values as exemplified by compound 38 (EC50 = 0.14 μM).

Small Molecules Targeting Human UDP-GlcNAc 2-Epimerase.

Uridine diphosphate N-acetylglucosamine 2-epimerase (GNE) is a key enzyme in the sialic acid biosynthesis pathway. Sialic acids are primarily terminal carbohydrates on glycans and play fundamental roles in health and disease. In search of effective GNE inhibitors not based on a carbohydrate scaffold, we performed a high-throughput screening campaign of 68,640 drug-like small molecules against recombinant GNE using a UDP detection assay. We validated nine of the primary actives with an orthogonal real-time NMR assay and verified their IC50 values in the low micromolar to nanomolar range manually. Stability and solubility studies revealed three compounds for further evaluation. Thermal shift assays, analytical size exclusion, and interferometric scattering microscopy demonstrated that the GNE inhibitors acted on the oligomeric state of the protein. Finally, hydrogen-deuterium exchange mass spectrometry (HDX-MS) revealed which sections of GNE were shifted upon the addition of the inhibitors. In summary, we have identified three small molecules as GNE inhibitors with high potency in vitro, which serve as promising candidates to modulate sialic acid biosynthesis in more complex systems.

Exploring the Binding Pathway of Novel Nonpeptidomimetic Plasmepsin V Inhibitors.

Predicting the interaction modes and binding affinities of virtual compound libraries is of great interest in drug development. It reduces the cost and time of lead compound identification and selection. Here we apply path-based metadynamics simulations to characterize the binding of potential inhibitors to the Plasmodium falciparum aspartic protease plasmepsin V (plm V), a validated antimalarial drug target that has a highly mobile binding site. The potential plm V binders were identified in a high-throughput virtual screening (HTVS) campaign and were experimentally verified in a fluorescence resonance energy transfer (FRET) assay. Our simulations allowed us to estimate compound binding energies and revealed relevant states along binding/unbinding pathways in atomistic resolution. We believe that the method described allows the prioritization of compounds for synthesis and enables rational structure-based drug design for targets that undergo considerable conformational changes upon inhibitor binding.

Acoustic droplet ejection facilitates cell-based high-throughput screenings using natural products

Natural Products (NPs) are one of the main sources for drug discovery. Many clinical drugs are NPs or NP-inspired compounds, and recently discovered New Chemical Entities (NCEs) of NPs are emerging as promising new drugs. High-Throughput Screening (HTS) of large sample sets or libraries has grown to be vital for the drug discovery field. Industrial-scale HTS of NP libraries can be limited due to the difficulties entailed in working with tiny extract volumes and the variability in viscosity of NP extracts. For these reasons, the implementation of new technologies to miniaturize different reagent volumes grows to be fundamental. Since Acoustic Droplet Ejection (ADE) emerged as a helpful tool in HTS campaigns for the transference of compound libraries. The aim of this work was to test the effectiveness of ADE for the dispensation of NP extract libraries in cell-based HTS assays.

Combining Molecular Quantum Mechanical Modeling and Machine Learning for Accelerated Reaction Screening and Discovery.

Molecular quantum mechanical modeling, accelerated by machine learning, has opened the door to high-throughput screening campaigns of complex properties, such as the activation energies of chemical reactions and absorption/emission spectra of materials and molecules; in silico. Here, we present an overview of the main principles, concepts, and design considerations involved in such hybrid computational quantum chemistry/machine learning screening workflows, with a special emphasis on some recent examples of their successful application. We end with a brief outlook of further advances that will benefit the field.

Molecular Dynamics as a Tool for Virtual Ligand Screening.

Rational drug design is essential for new drugs to emerge, especially when the structure of a target protein or nucleic acid is known. To that purpose, high-throughput virtual ligand screening campaigns aim at discovering computationally new binding molecules or fragments to modulate particular biomolecular interactions or biological activities, related to a disease process. The structure-based virtual ligand screening process primarily relies on docking methods which allow predicting the binding of a molecule to a biological target structure with a correct conformation and the best possible affinity. The docking method itself is not sufficient as it suffers from several and crucial limitations (lack of full protein flexibility information, no solvation and ion effects, poor scoring functions, and unreliable molecular affinity estimation).At the interface of computer techniques and drug discovery, molecular dynamics (MD) allows introducing protein flexibility before or after a docking protocol, refining the structure of protein-drug complexes in the presence of water, ions, and even in membrane-like environments, describing more precisely the temporal evolution of the biological complex and ranking these complexes with more accurate binding energy calculations. In this chapter, we describe the up-to-date MD, which plays the role of supporting tools in the virtual ligand screening (VS) process.Without a doubt, using docking in combination with MD is an attractive approach in structure-based drug discovery protocols nowadays. It has proved its efficiency through many examples in the literature and is a powerful method to significantly reduce the amount of required wet experimentations (Tarcsay et al, J Chem Inf Model 53:2990-2999, 2013; Barakat et al, PLoS One 7:e51329, 2012; De Vivo et al, J Med Chem 59:4035-4061, 2016; Durrant, McCammon, BMC Biol 9:71-79, 2011; Galeazzi, Curr Comput Aided Drug Des 5:225-240, 2009; Hospital et al, Adv Appl Bioinforma Chem 8:37-47, 2015; Jiang et al, Molecules 20:12769-12786, 2015; Kundu et al, J Mol Graph Model 61:160-174, 2015; Mirza et al, J Mol Graph Model 66:99-107, 2016; Moroy et al, Future Med Chem 7:2317-2331, 2015; Naresh et al, J Mol Graph Model 61:272-280, 2015; Nichols et al, J Chem Inf Model 51:1439-1446, 2011; Nichols et al, Methods Mol Biol 819:93-103, 2012; Okimoto et al, PLoS Comput Biol 5:e1000528, 2009; Rodriguez-Bussey et al, Biopolymers 105:35-42, 2016; Sliwoski et al, Pharmacol Rev 66:334-395, 2014).

Comparative Performance of High-Throughput Methods for Protein pKa Predictions.

The medically relevant field of protein-based therapeutics has triggered a demand for protein engineering in different pH environments of biological relevance. In silico engineering workflows typically employ high-throughput screening campaigns that require evaluating large sets of protein residues and point mutations by fast yet accurate computational algorithms. While several high-throughput pKa prediction methods exist, their accuracies are unclear due to the lack of a current comprehensive benchmarking. Here, seven fast, efficient, and accessible approaches including PROPKA3, DeepKa, PKAI, PKAI+, DelPhiPKa, MCCE2, and H++ were systematically tested on a nonredundant subset of 408 measured protein residue pKa shifts from the pKa database (PKAD). While no method outperformed the null hypotheses with confidence, as illustrated by statistical bootstrapping, DeepKa, PKAI+, PROPKA3, and H++ had utility. More specifically, DeepKa consistently performed well in tests across multiple and individual amino acid residue types, as reflected by lower errors, higher correlations, and improved classifications. Arithmetic averaging of the best empirical predictors into simple consensuses improved overall transferability and accuracy up to a root-mean-square error of 0.76 pKa units and a correlation coefficient (R2) of 0.45 to experimental pKa shifts. This analysis should provide a basis for further methodological developments and guide future applications, which require embedding of computationally inexpensive pKa prediction methods, such as the optimization of antibodies for pH-dependent antigen binding.

Label-free high-throughput screening via acoustic ejection mass spectrometry put into practice.

Acoustic droplet ejection-open port interface-mass spectrometry (ADE-OPI-MS) is a novel label-free analytical technique, promising to become a versatile readout for high-throughput screening (HTS) applications. The recent introduction of ADE-OPI-MS devices to the laboratory equipment market, paired with their compatibility with laboratory automation platforms, should facilitate the adoption of this technology by a broader community. Towards this goal, instrument robustness in the context of HTS campaigns - where up to millions of samples in complex matrices are tested in a short time frame - represents a major challenge, which explains the absence of detailed literature reports on this subject. Here, we present the results of our first fully automated HTS campaign, based on the ADE-OPI-MS technology, aiming to identify inhibitors of a metabolic enzyme in a >1 million compound library. The report encompasses the assay development and validation steps, as well as the adaptation for HTS requirements, where refinement of the capillary cleaning concept was crucial for final success. Altogether, our study unequivocally demonstrates the applicability of the ADE-OPI-MS technology for HTS-based drug discovery.

Mathematical relationships between control group variability and assay quality metrics.

Assay quality metrics have been used in various high-throughput screening (HTS) campaigns to indicate assay quality. Z'-factor has become one of the most widely used metrics, along with other metrics such as standardised mean difference (SSMD). In using these metrics, it is important to understand how these metrics can be impacted by the separation between control groups (indicated by the HZ ratio) and the coefficient of variation (CV) within each control group. In this paper, several mathematical equations have been derived to understand the relationship between assay quality metrics (such as Z'-factor and SSMD) and control group datasets (summarised by CV and HZ). These equations increase our understanding of the factors that improve assay quality metrics, thus providing a quantitative means to visualise how affecting control groups can impact assay quality metrics.

Structure-based computational design of novel covalent binders for the treatment of sickle cell disease

The quest in finding an everlasting panacea to the pernicious impact of sickle cell disease (SCD) in the society hit a turn of success since the recent discovery of a small molecule reversible covalent inhibitor, Voxelotor. A drug that primarily promotes the stability of oxygenated hemoglobin and inhibit the polymerization of HbS by enhancing hemoglobin's affinity for oxygen has opened a new frontier in drug discovery and development. Despite eminent efforts made to reproduce small molecules with better therapeutic targets, none has been successful. To this end, we employed the use of structure-based computational techniques with emphasis on the electrophilic warhead group of Voxelotor to harness novel covalent binders that could elicit better therapeutic response against HbS. The PubChem database and DataWarrior software were used to design random molecules using Voxelotor's electrophilic functionality. Following the compilation of these chemical entities, a high-throughput covalent docking-based virtual screening campaign was conducted which revealed three (Compound_166, Compound_2301, and Compound_2335) putative druglike candidates with higher baseline energy value compared to the standard drug. Subsequently, in silico ADMET profiling was carried out to evaluate their pharmacokinetics and pharmacodynamics properties, and their stability was evaluated for 1 μs (1 μs) using molecular dynamics simulation. Finally, to prioritize these compounds for further development in drug discovery, MM/PBSA calculations was employed to evaluate their molecular interactions and solvation energy within the HbS protein. Despite the admirable druglike and stability properties of these compounds, further experimental validations are required to establish their preclinical relevance for drug development.

MF-PCBA: Multifidelity High-Throughput Screening Benchmarks for Drug Discovery and Machine Learning.

High-throughput screening (HTS), as one of the key techniques in drug discovery, is frequently used to identify promising drug candidates in a largely automated and cost-effective way. One of the necessary conditions for successful HTS campaigns is a large and diverse compound library, enabling hundreds of thousands of activity measurements per project. Such collections of data hold great promise for computational and experimental drug discovery efforts, especially when leveraged in combination with modern deep learning techniques, and can potentially lead to improved drug activity predictions and cheaper and more effective experimental design. However, existing collections of machine-learning-ready public datasets do not exploit the multiple data modalities present in real-world HTS projects. Thus, the largest fraction of experimental measurements, corresponding to hundreds of thousands of "noisy" activity values from primary screening, are effectively ignored in the majority of machine learning models of HTS data. To address these limitations, we introduce Multifidelity PubChem BioAssay (MF-PCBA), a curated collection of 60 datasets that includes two data modalities for each dataset, corresponding to primary and confirmatory screening, an aspect that we call multifidelity. Multifidelity data accurately reflect real-world HTS conventions and present a new, challenging task for machine learning: the integration of low- and high-fidelity measurements through molecular representation learning, taking into account the orders-of-magnitude difference in size between the primary and confirmatory screens. Here we detail the steps taken to assemble MF-PCBA in terms of data acquisition from PubChem and the filtering steps required to curate the raw data. We also provide an evaluation of a recent deep-learning-based method for multifidelity integration across the introduced datasets, demonstrating the benefit of leveraging all HTS modalities, and a discussion in terms of the roughness of the molecular activity landscape. In total, MF-PCBA contains over 16.6 million unique molecule-protein interactions. The datasets can be easily assembled by using the source code available at https://github.com/davidbuterez/mf-pcba.

Methyl and Fluorine Effects in Novel Orally Bioavailable Keap1-Nrf2 PPI Inhibitor.

Oxidative stress is one of the causes of progression of chronic kidney disease (CKD). Activation of the antioxidant protein regulator Nrf2 by inhibition of the Keap1-Nrf2 protein-protein interaction (PPI) is of interest as a potential treatment for CKD. We report the identification of the novel and weak PPI inhibitor 7 with good physical properties by a high throughput screening (HTS) campaign, followed by structural and computational analysis. The installation of only methyl and fluorine groups successfully provided the lead compound 25, which showed more than 400-fold stronger activity. Furthermore, these dramatic substituent effects can be explained by the analysis of using isothermal titration calorimetry (ITC). Thus, the resulting 25, which exhibited high oral absorption and durability, would be a CKD therapeutic agent because of the dose-dependent manner for up-regulation of the antioxidant protein heme oxigenase-1 (HO-1) in rat kidneys.