High-content phenotypic and pathway profiling to advance drug discovery in diseases of unmet need

Artificial intelligence for high content imaging in drug discovery

Artificial intelligence (AI) and high-content imaging (HCI) are contributing to advancements in drug discovery, propelled by the recent progress in deep neural networks. This review highlights AI's role in analysis of HCI data from fixed and live-cell imaging, enabling novel label-free and multi-channel fluorescent screening methods, and improving compound profiling. HCI experiments are rapid and cost-effective, facilitating large data set accumulation for AI model training. However, the success of AI in drug discovery also depends on high-quality data, reproducible experiments, and robust validation to ensure model performance. Despite challenges like the need for annotated compounds and managing vast image data, AI's potential in phenotypic screening and drug profiling is significant. Future improvements in AI, including increased interpretability and integration of multiple modalities, are expected to solidify AI and HCI's role in drug discovery.

A Decade in a Systematic Review: The Evolution and Impact of Cell Painting

High-content image-based assays have fueled significant discoveries in the life sciences in the past decade (2013-2023), including novel insights into disease etiology, mechanism of action, new therapeutics, and toxicology predictions. Here, we systematically review the substantial methodological advancements and applications of Cell Painting. Advancements include improvements in the Cell Painting protocol, assay adaptations for different types of perturbations and applications, and improved methodologies for feature extraction, quality control, and batch effect correction. Moreover, machine learning methods recently surpassed classical approaches in their ability to extract biologically useful information from Cell Painting images. Cell Painting data have been used alone or in combination with other - omics data to decipher the mechanism of action of a compound, its toxicity profile, and many other biological effects. Overall, key methodological advances have expanded Cell Painting's ability to capture cellular responses to various perturbations. Future advances will likely lie in advancing computational and experimental techniques, developing new publicly available datasets, and integrating them with other high-content data types.

Multipathway regulation induced by 4-(phenylsulfonyl)morpholine derivatives against triple-negative breast cancer

Phenotypic drug discovery (PDD) is an effective drug discovery approach by observation of therapeutic effects on disease phenotypes, especially in complex disease systems. Triple-negative breast cancer (TNBC) is composed of several complex disease features, including high tumor heterogeneity, high invasive and metastatic potential, and a lack of effective therapeutic targets. Therefore, identifying effective and novel agents through PDD is a current trend in TNBC drug development. In this study, 23 novel small molecules were synthesized using 4-(phenylsulfonyl)morpholine as a pharmacophore. Among these derivatives, GL24 (4m) exhibited the lowest half-maximal inhibitory concentration value (0.90 µM) in MDA-MB-231 cells. To investigate the tumor-suppressive mechanisms of GL24, transcriptomic analyses were used to detect the perturbation for gene expression upon GL24 treatment. Followed by gene ontology (GO) analysis, gene set enrichment analysis (GSEA), and Kyoto Encyclopedia of Genes and Genomes (KEGG) analysis, multiple ER stress-dependent tumor suppressive signals were identified, such as unfolded protein response (UPR), p53 pathway, G2/M checkpoint, and E2F targets. Most of the identified pathways triggered by GL24 eventually led to cell-cycle arrest and then to apoptosis. In summary, we developed a novel 4-(phenylsulfonyl)morpholine derivative GL24 with a strong potential for inhibiting TNBC cell growth through ER stress-dependent tumor suppressive signals.

Expanding the Chemical Space of Transforming Growth Factor-β (TGFβ) Receptor Type II Degraders with 3,4-Disubstituted Indole Derivatives

The TGFβ type II receptor (TβRII) is a central player in all TGFβ signaling downstream events, has been linked to cancer progression, and thus, has emerged as an auspicious anti-TGFβ strategy. Especially its targeted degradation presents an excellent goal for effective TGFβ pathway inhibition. Here, cellular structure-activity relationship (SAR) data from the TβRII degrader chemotype 1 was successfully transformed into predictive ligand-based pharmacophore models that allowed scaffold hopping. Two distinct 3,4-disubstituted indoles were identified from virtual screening: tetrahydro-4-oxo-indole 2 and indole-3-acetate 3. Design, synthesis, and screening of focused amide libraries confirmed 2r and 3n as potent TGFβ inhibitors. They were validated to fully recapitulate the ability of 1 to selectively degrade TβRII, without affecting TβRI. Consequently, 2r and 3n efficiently blocked endothelial-to-mesenchymal transition and cell migration in different cancer cell lines while not perturbing the microtubule network. Hence, 2 and 3 present novel TβRII degrader chemotypes that will (1) aid target deconvolution efforts and (2) accelerate proof-of-concept studies for small-molecule-driven TβRII degradation in vivo.

In vitro models for neuropathic pain phenotypic screening in brain therapeutics

The discovery of brain therapeutics faces a significant challenge due to the low translatability of preclinical results into clinical success. To address this gap, several efforts have been made to obtain more translatable neuronal models for phenotypic screening. These models allow the selection of active compounds without predetermined knowledge of drug targets. In this review, we present an overview of various existing models within the field, examining their strengths and limitations, particularly in the context of neuropathic pain research. We illustrate the usefulness of these models through a comparative review in three crucial areas: i) the development of novel phenotypic screening strategies specifically for neuropathic pain, ii) the validation of the models for both primary and secondary screening assays, and iii) the use of the models in target deconvolution processes.

Target Prediction by Multiple Virtual Screenings: Analyzing the SARS-CoV-2 Phenotypic Screening by the Docking Simulations Submitted to the MEDIATE Initiative

Phenotypic screenings are usually combined with deconvolution techniques to characterize the mechanism of action for the retrieved hits. These studies can be supported by various computational analyses, although docking simulations are rarely employed. The present study aims to assess if multiple docking calculations can prove successful in target prediction. In detail, the docking simulations submitted to the MEDIATE initiative are utilized to predict the viral targets involved in the hits retrieved by a recently published cytopathic screening. Multiple docking results are combined by the EFO approach to develop target-specific consensus models. The combination of multiple docking simulations enhances the performances of the developed consensus models (average increases in EF1% value of 40% and 25% when combining three and two docking runs, respectively). These models are able to propose reliable targets for about half of the retrieved hits (31 out of 59). Thus, the study emphasizes that docking simulations might be effective in target identification and provide a convincing validation for the collaborative strategies that inspire the MEDIATE initiative. Disappointingly, cross-target and cross-program correlations suggest that common scoring functions are not specific enough for the simulated target.

Small-Molecule Probes as Pharmacological Tools for the Bone Morphogenetic Protein Signaling Pathway

The bone morphogenetic protein (BMP) pathway is highly conserved and plays central roles in health and disease. The quality and quantity of its signaling outputs are regulated at multiple levels, offering pharmacological options for targeted modulation. Both target-centric and phenotypic drug discovery (PDD) approaches were applied to identify small-molecule BMP inhibitors and stimulators. In this Review, we accumulated and systematically classified the different reported chemotypes based on their targets as well as modes-of-action, and herein we illustrate the discovery history of selected candidates. A comprehensive summary of available biochemical, cellular, and in vivo activities is provided for the most relevant BMP modulators, along with recommendations on their preferred use as chemical probes to study BMP-related (patho)physiological processes. There are a number of high-quality probes used as BMP inhibitors that potently and selectively interrogate the kinase activities of distinct type I (16 chemotypes available) and type II receptors (3 chemotypes available). In contrast, only a few high-quality BMP stimulator modalities have been introduced to the field due to a lack of profound target knowledge. FK506-derived macrolides such as calcineurin-sparing FKBP12 inhibitors currently represent the best-characterized chemical tools for direct activation of BMP-SMAD signaling at the receptor level. However, several PDD campaigns succeeded in expanding the druggable space of BMP stimulators. Albeit the majority of them do not entirely fulfill the strict chemical probe criteria, many chemotypes exhibit unique and unrecognized mechanisms as pathway potentiators or synergizers, serving as valuable pharmacological tools for BMP perturbation.

Discovering the mechanism of action of drugs with a sparse explainable network

BACKGROUND

Although Deep Neural Networks (DDNs) have been successful in predicting the efficacy of cancer drugs, the lack of explainability in their decision-making process is a significant challenge. Previous research proposed mimicking the Gene Ontology structure to allow for interpretation of each neuron in the network. However, these previous approaches require huge amount of GPU resources and hinder its extension to genome-wide models.

METHODS

We developed SparseGO, a sparse and interpretable neural network, for predicting drug response in cancer cell lines and their Mechanism of Action (MoA). To ensure model generalization, we trained it on multiple datasets and evaluated its performance using three cross-validation schemes. Its efficiency allows it to be used with gene expression. In addition, SparseGO integrates an eXplainable Artificial Intelligence (XAI) technique, DeepLIFT, with Support Vector Machines to computationally discover the MoA of drugs.

FINDINGS

SparseGO's sparse implementation significantly reduced GPU memory usage and training speed compared to other methods, allowing it to process gene expression instead of mutations as input data. SparseGO using expression improved the accuracy and enabled its use on drug repositioning. Furthermore, gene expression allows the prediction of MoA using 265 drugs to train it. It was validated on understudied drugs such as parbendazole and PD153035.

INTERPRETATION

SparseGO is an effective XAI method for predicting, but more importantly, understanding drug response.

FUNDING

The Accelerator Award Programme funded by Cancer Research UK [C355/A26819], Fundación Científica de la AECC and Fondazione AIRC, Project PIBA_2020_1_0055 funded by the Basque Government and the Synlethal Project (RETOS Investigacion, Spanish Government).

Discovering cellular programs of intrinsic and extrinsic drivers of metabolic traits using LipocyteProfiler

A primary obstacle in translating genetic associations with disease into therapeutic strategies is elucidating the cellular programs affected by genetic risk variants and effector genes. Here, we introduce LipocyteProfiler, a cardiometabolic-disease-oriented high-content image-based profiling tool that enables evaluation of thousands of morphological and cellular profiles that can be systematically linked to genes and genetic variants relevant to cardiometabolic disease. We show that LipocyteProfiler allows surveillance of diverse cellular programs by generating rich context- and process-specific cellular profiles across hepatocyte and adipocyte cell-state transitions. We use LipocyteProfiler to identify known and novel cellular mechanisms altered by polygenic risk of metabolic disease, including insulin resistance, fat distribution, and the polygenic contribution to lipodystrophy. LipocyteProfiler paves the way for large-scale forward and reverse deep phenotypic profiling in lipocytes and provides a framework for the unbiased identification of causal relationships between genetic variants and cellular programs relevant to human disease.

Discovery of senolytics using machine learning

Cellular senescence is a stress response involved in ageing and diverse disease processes including cancer, type-2 diabetes, osteoarthritis and viral infection. Despite growing interest in targeted elimination of senescent cells, only few senolytics are known due to the lack of well-characterised molecular targets. Here, we report the discovery of three senolytics using cost-effective machine learning algorithms trained solely on published data. We computationally screened various chemical libraries and validated the senolytic action of ginkgetin, periplocin and oleandrin in human cell lines under various modalities of senescence. The compounds have potency comparable to known senolytics, and we show that oleandrin has improved potency over its target as compared to best-in-class alternatives. Our approach led to several hundred-fold reduction in drug screening costs and demonstrates that artificial intelligence can take maximum advantage of small and heterogeneous drug screening data, paving the way for new open science approaches to early-stage drug discovery.

Transcriptional Profiling of a Patient-Matched Cohort of Glioblastoma (IDH-Wildtype) for Therapeutic Target and Repurposing Drug Identification

Glioblastoma (GBM) is the most prevalent and aggressive adult brain tumor. Despite multi-modal therapies, GBM recurs, and patients have poor survival (~14 months). Resistance to therapy may originate from a subpopulation of tumor cells identified as glioma-stem cells (GSC), and new treatments are urgently needed to target these. The biology underpinning GBM recurrence was investigated using whole transcriptome profiling of patient-matched initial and recurrent GBM (recGBM). Differential expression analysis identified 147 significant probes. In total, 24 genes were validated using expression data from four public cohorts and the literature. Functional analyses revealed that transcriptional changes to recGBM were dominated by angiogenesis and immune-related processes. The role of MHC class II proteins in antigen presentation and the differentiation, proliferation, and infiltration of immune cells was enriched. These results suggest recGBM would benefit from immunotherapies. The altered gene signature was further analyzed in a connectivity mapping analysis with QUADrATiC software to identify FDA-approved repurposing drugs. Top-ranking target compounds that may be effective against GSC and GBM recurrence were rosiglitazone, nizatidine, pantoprazole, and tolmetin. Our translational bioinformatics pipeline provides an approach to identify target compounds for repurposing that may add clinical benefit in addition to standard therapies against resistant cancers such as GBM.

MAC-Seq: Coupling Low-Cost, High-Throughput RNA-Seq with Image-Based Phenotypic Screening in 2D and 3D Cell Models

Transcriptomic profiling has fundamentally influenced our understanding of cancer pathophysiology and response to therapeutic intervention and has become a relatively routine approach. However, standard protocols are usually low-throughput, single-plex assays and costs are still quite prohibitive. With the evolving complexity of in vitro cell model systems, there is a need for resource-efficient high-throughput approaches that can support detailed time-course analytics, accommodate limited sample availability, and provide the capacity to correlate phenotype to genotype at scale. MAC-seq (multiplexed analysis of cells) is a low-cost, ultrahigh-throughput RNA-seq workflow in plate format to measure cell perturbations and is compatible with high-throughput imaging. Here we describe the steps to perform MAC-seq in 384-well format and apply it to 2D and 3D cell cultures. On average, our experimental conditions identified over ten thousand expressed genes per well when sequenced to a depth of one million reads. We discuss technical aspects, make suggestions on experimental design, and document critical operational procedures. Our protocol highlights the potential to couple MAC-seq with high-throughput screening applications including cell phenotyping using high-content cell imaging.

Scaffold-based three-dimensional cell model of pancreatic cancer is more suitable than scaffold-free three-dimensional cell model of pancreatic cancer for drug discovery

Pancreatic cancer is one of the deadliest malignancies. Three-dimensional (3D) pancreatic cancer cell models for drug screening have been established to improve treatment for pancreatic cancer. However, few studies focus on different drug responses and drug-related molecular mechanisms in various types of 3D cell models. In this study, we constructed 3D scaffold-free cell models and 3D scaffold-based cell models of pancreatic cancer, evaluated chemotherapeutic drug responses in different 3D models, assessed clinical relevance of the models, and investigated molecular mechanisms of chemoresistance and drug pathways in different 3D models. Both types of 3D models showed resistance to chemotherapeutic drugs, and scaffold-based pancreatic cancer models could better reflect in vivo drug efficacy than 2D and scaffold-free pancreatic cancer models did. Increased cell adhesion, extracellular matrix (ECM) synthesis and drug transport were essential for drug resistance in 3D models, and anti-apoptosis might contribute to extreme chemoresistance in scaffold-free models. Moreover, scaffold-based pancreatic cancer models were more suitable than scaffold-free models for drug pathway research.

Virtual screening for small-molecule pathway regulators by image-profile matching

Identifying the chemical regulators of biological pathways is a time-consuming bottleneck in developing therapeutics and research compounds. Typically, thousands to millions of candidate small molecules are tested in target-based biochemical screens or phenotypic cell-based screens, both expensive experiments customized to each disease. Here, our uncustomized, virtual, profile-based screening approach instead identifies compounds that match to pathways based on the phenotypic information in public cell image data, created using the Cell Painting assay. Our straightforward correlation-based computational strategy retrospectively uncovered the expected, known small-molecule regulators for 32% of positive-control gene queries. In prospective, discovery mode, we efficiently identified new compounds related to three query genes and validated them in subsequent gene-relevant assays, including compounds that phenocopy or pheno-oppose YAP1 overexpression and kill a Yap1-dependent sarcoma cell line. This image-profile-based approach could replace many customized labor- and resource-intensive screens and accelerate the discovery of biologically and therapeutically useful compounds.

High-throughput profiling of histone post-translational modifications and chromatin modifying proteins by reverse phase protein array

Epigenetic variation plays a significant role in normal development and human diseases including cancer, in part through post-translational modifications (PTMs) of histones. Identification and profiling of changes in histone PTMs, and in proteins regulating PTMs, are crucial to understanding diseases, and for discovery of epigenetic therapeutic agents. In this study, we have adapted and validated an antibody-based reverse phase protein array (RPPA) platform for profiling 20 histone PTMs and expression of 40 proteins that modify histones and other epigenomic regulators. The specificity of the RPPA assay for histone PTMs was validated with synthetic peptides corresponding to histone PTMs and by detection of histone PTM changes in response to inhibitors of histone modifier proteins in cell cultures. The useful application of the RPPA platform was demonstrated with two models: induction of pluripotent stem cells and a mouse mammary tumor progression model. Described here is a robust platform that includes a rapid microscale method for histone isolation and partially automated workflows for analysis of histone PTMs and histone modifiers that can be performed in a high-throughput manner with hundreds of samples. This RPPA platform has potential for translational applications through the discovery and validation of epigenetic states as therapeutic targets and biomarkers. SIGNIFICANCE: Our study has established an antibody-based reverse phase protein array platform for global profiling of a wide range of post-translational modifications of histones and histone modifier proteins. The high-throughput platform provides comprehensive analyses of epigenetics for biological research and disease studies and may serve as screening assay for diagnostic purpose or therapy development.

Machine Learning Enables Accurate and Rapid Prediction of Active Molecules Against Breast Cancer Cells

Breast cancer (BC) has surpassed lung cancer as the most frequently occurring cancer, and it is the leading cause of cancer-related death in women. Therefore, there is an urgent need to discover or design new drug candidates for BC treatment. In this study, we first collected a series of structurally diverse datasets consisting of 33,757 active and 21,152 inactive compounds for 13 breast cancer cell lines and one normal breast cell line commonly used in in vitro antiproliferative assays. Predictive models were then developed using five conventional machine learning algorithms, including naïve Bayesian, support vector machine, k-Nearest Neighbors, random forest, and extreme gradient boosting, as well as five deep learning algorithms, including deep neural networks, graph convolutional networks, graph attention network, message passing neural networks, and Attentive FP. A total of 476 single models and 112 fusion models were constructed based on three types of molecular representations including molecular descriptors, fingerprints, and graphs. The evaluation results demonstrate that the best model for each BC cell subtype can achieve high predictive accuracy for the test sets with AUC values of 0.689–0.993. Moreover, important structural fragments related to BC cell inhibition were identified and interpreted. To facilitate the use of the model, an online webserver called ChemBC (http://chembc.idruglab.cn/) and its local version software (https://github.com/idruglab/ChemBC) were developed to predict whether compounds have potential inhibitory activity against BC cells.

Type III secretion system effector subnetworks elicit distinct host immune responses to infection

Citrobacter rodentium, a natural mouse pathogen which colonises the colon of immuno-competent mice, provides a robust model for interrogating host-pathogen-microbiota interactions in vivo. This model has been key to providing new insights into local host responses to enteric infection, including changes in intestinal epithelial cell immunometabolism and mucosal immunity. C. rodentium injects 31 bacterial effectors into epithelial cells via a type III secretion system (T3SS). Recently, these effectors were shown to be able to form multiple intracellular subnetworks which can withstand significant contractions whilst maintaining virulence. Here we highlight recent advances in understanding gut mucosal responses to infection and effector biology, as well as potential uses for artificial intelligence (AI) in understanding infectious disease and speculate on the role of T3SS effector networks in host adaption.

BowSaw: Inferring Higher-Order Trait Interactions Associated With Complex Biological Phenotypes

Machine learning is helping the interpretation of biological complexity by enabling the inference and classification of cellular, organismal and ecological phenotypes based on large datasets, e.g., from genomic, transcriptomic and metagenomic analyses. A number of available algorithms can help search these datasets to uncover patterns associated with specific traits, including disease-related attributes. While, in many instances, treating an algorithm as a black box is sufficient, it is interesting to pursue an enhanced understanding of how system variables end up contributing to a specific output, as an avenue toward new mechanistic insight. Here we address this challenge through a suite of algorithms, named BowSaw, which takes advantage of the structure of a trained random forest algorithm to identify combinations of variables (“rules”) frequently used for classification. We first apply BowSaw to a simulated dataset and show that the algorithm can accurately recover the sets of variables used to generate the phenotypes through complex Boolean rules, even under challenging noise levels. We next apply our method to data from the integrative Human Microbiome Project and find previously unreported high-order combinations of microbial taxa putatively associated with Crohn’s disease. By leveraging the structure of trees within a random forest, BowSaw provides a new way of using decision trees to generate testable biological hypotheses.