A community challenge for a pancancer drug mechanism of action inference from perturbational profile data

Assessing the impact of transcriptomics data analysis pipelines on downstream functional enrichment results

Transcriptomics is widely used to assess the state of biological systems. There are many tools for the different steps, such as normalization, differential expression, and enrichment. While numerous studies have examined the impact of method choices on differential expression results, little attention has been paid to their effects on further downstream functional analysis, which typically provides the basis for interpretation and follow-up experiments. To address this, we introduce FLOP, a comprehensive nextflow-based workflow combining methods to perform end-to-end analyses of transcriptomics data. We illustrate FLOP on datasets ranging from end-stage heart failure patients to cancer cell lines. We discovered effects not noticeable at the gene-level, and observed that not filtering the data had the highest impact on the correlation between pipelines in the gene set space. Moreover, we performed three benchmarks to evaluate the 12 pipelines included in FLOP, and confirmed that filtering is essential in scenarios of expected moderate-to-low biological signal. Overall, our results underscore the impact of carefully evaluating the consequences of the choice of preprocessing methods on downstream enrichment analyses. We envision FLOP as a valuable tool to measure the robustness of functional analyses, ultimately leading to more reliable and conclusive biological findings.

CODEX: COunterfactual Deep learning for the in silico EXploration of cancer cell line perturbations

MOTIVATION

High-throughput screens (HTS) provide a powerful tool to decipher the causal effects of chemical and genetic perturbations on cancer cell lines. Their ability to evaluate a wide spectrum of interventions, from single drugs to intricate drug combinations and CRISPR-interference, has established them as an invaluable resource for the development of novel therapeutic approaches. Nevertheless, the combinatorial complexity of potential interventions makes a comprehensive exploration intractable. Hence, prioritizing interventions for further experimental investigation becomes of utmost importance.

RESULTS

We propose CODEX (COunterfactual Deep learning for the in silico EXploration of cancer cell line perturbations) as a general framework for the causal modeling of HTS data, linking perturbations to their downstream consequences. CODEX relies on a stringent causal modeling strategy based on counterfactual reasoning. As such, CODEX predicts drug-specific cellular responses, comprising cell survival and molecular alterations, and facilitates the in silico exploration of drug combinations. This is achieved for both bulk and single-cell HTS. We further show that CODEX provides a rationale to explore complex genetic modifications from CRISPR-interference in silico in single cells.

AVAILABILITY AND IMPLEMENTATION

Our implementation of CODEX is publicly available at https://github.com/sschrod/CODEX. All data used in this article are publicly available.

Deep representation learning of chemical-induced transcriptional profile for phenotype-based drug discovery

Artificial intelligence transforms drug discovery, with phenotype-based approaches emerging as a promising alternative to target-based methods, overcoming limitations like lack of well-defined targets. While chemical-induced transcriptional profiles offer a comprehensive view of drug mechanisms, inherent noise often obscures the true signal, hindering their potential for meaningful insights. Here, we highlight the development of TranSiGen, a deep generative model employing self-supervised representation learning. TranSiGen analyzes basal cell gene expression and molecular structures to reconstruct chemical-induced transcriptional profiles with high accuracy. By capturing both cellular and compound information, TranSiGen-derived representations demonstrate efficacy in diverse downstream tasks like ligand-based virtual screening, drug response prediction, and phenotype-based drug repurposing. Notably, in vitro validation of TranSiGen's application in pancreatic cancer drug discovery highlights its potential for identifying effective compounds. We envisage that integrating TranSiGen into the drug discovery and mechanism research holds significant promise for advancing biomedicine.

Targeting neddylation and sumoylation in chemoresistant triple negative breast cancer

Triple negative breast cancer (TNBC) accounts for 15-20% of breast cancer cases in the United States. Systemic neoadjuvant chemotherapy (NACT), with or without immunotherapy, is the current standard of care for patients with early-stage TNBC. However, up to 70% of TNBC patients have significant residual disease once NACT is completed, which is associated with a high risk of developing recurrence within two to three years of surgical resection. To identify targetable vulnerabilities in chemoresistant TNBC, we generated longitudinal patient-derived xenograft (PDX) models from TNBC tumors before and after patients received NACT. We then compiled transcriptomes and drug response profiles for all models. Transcriptomic analysis identified the enrichment of aberrant protein homeostasis pathways in models from post-NACT tumors relative to pre-NACT tumors. This observation correlated with increased sensitivity in vitro to inhibitors targeting the proteasome, heat shock proteins, and neddylation pathways. Pevonedistat, a drug annotated as a NEDD8-activating enzyme (NAE) inhibitor, was prioritized for validation in vivo and demonstrated efficacy as a single agent in multiple PDX models of TNBC. Pharmacotranscriptomic analysis identified a pathway-level correlation between pevonedistat activity and post-translational modification (PTM) machinery, particularly involving neddylation and sumoylation targets. Elevated levels of both NEDD8 and SUMO1 were observed in models exhibiting a favorable response to pevonedistat compared to those with a less favorable response in vivo. Moreover, a correlation emerged between the expression of neddylation-regulated pathways and tumor response to pevonedistat, indicating that targeting these PTM pathways may prove effective in combating chemoresistant TNBC.

Inference of drug off-target effects on cellular signaling using interactome-based deep learning

Many diseases emerge from dysregulated cellular signaling, and drugs are often designed to target specific signaling proteins. Off-target effects are, however, common and may ultimately result in failed clinical trials. Here we develop a computer model of the cell's transcriptional response to drugs for improved understanding of their mechanisms of action. The model is based on ensembles of artificial neural networks and simultaneously infers drug-target interactions and their downstream effects on intracellular signaling. With this, it predicts transcription factors' activities, while recovering known drug-target interactions and inferring many new ones, which we validate with an independent dataset. As a case study, we analyze the effects of the drug Lestaurtinib on downstream signaling. Alongside its intended target, FLT3, the model predicts an inhibition of CDK2 that enhances the downregulation of the cell cycle-critical transcription factor FOXM1. Our approach can therefore enhance our understanding of drug signaling for therapeutic design.

Elucidation and Pharmacologic Targeting of Master Regulator Dependencies in Coexisting Diffuse Midline Glioma Subpopulations

Diffuse Midline Gliomas (DMGs) are universally fatal, primarily pediatric malignancies affecting the midline structures of the central nervous system. Despite decades of clinical trials, treatment remains limited to palliative radiation therapy. A major challenge is the coexistence of molecularly distinct malignant cell states with potentially orthogonal drug sensitivities. To address this challenge, we leveraged established network-based methodologies to elucidate Master Regulator (MR) proteins representing mechanistic, non-oncogene dependencies of seven coexisting subpopulations identified by single-cell analysis-whose enrichment in essential genes was validated by pooled CRISPR/Cas9 screens. Perturbational profiles of 372 clinically relevant drugs helped identify those able to invert the activity of subpopulation-specific MRs for follow-up in vivo validation. While individual drugs predicted to target individual subpopulations-including avapritinib, larotrectinib, and ruxolitinib-produced only modest tumor growth reduction in orthotopic models, systemic co-administration induced significant survival extension, making this approach a valuable contribution to the rational design of combination therapy.

Cancer Mutations Converge on a Collection of Protein Assemblies to Predict Resistance to Replication Stress

Rapid proliferation is a hallmark of cancer associated with sensitivity to therapeutics that cause DNA replication stress (RS). Many tumors exhibit drug resistance, however, via molecular pathways that are incompletely understood. Here, we develop an ensemble of predictive models that elucidate how cancer mutations impact the response to common RS-inducing (RSi) agents. The models implement recent advances in deep learning to facilitate multidrug prediction and mechanistic interpretation. Initial studies in tumor cells identify 41 molecular assemblies that integrate alterations in hundreds of genes for accurate drug response prediction. These cover roles in transcription, repair, cell-cycle checkpoints, and growth signaling, of which 30 are shown by loss-of-function genetic screens to regulate drug sensitivity or replication restart. The model translates to cisplatin-treated cervical cancer patients, highlighting an RTK-JAK-STAT assembly governing resistance. This study defines a compendium of mechanisms by which mutations affect therapeutic responses, with implications for precision medicine.

SIGNIFICANCE

Zhao and colleagues use recent advances in machine learning to study the effects of tumor mutations on the response to common therapeutics that cause RS. The resulting predictive models integrate numerous genetic alterations distributed across a constellation of molecular assemblies, facilitating a quantitative and interpretable assessment of drug response. This article is featured in Selected Articles from This Issue, p. 384.

The Status Quo of Pharmacogenomics of Tyrosine Kinase Inhibitors in Precision Oncology: A Bibliometric Analysis of the Literature

Precision oncology and pharmacogenomics (PGx) intersect in their overarching goal to institute the right treatment for the right patient. However, the translation of these innovations into clinical practice is still lagging behind. Therefore, this study aimed to analyze the current state of research and to predict the future directions of applied PGx in the field of precision oncology as represented by the targeted therapy class of tyrosine kinase inhibitors (TKIs). Advanced bibliometric and scientometric analyses of the literature were performed. The Scopus database was used for the search, and articles published between 2001 and 2023 were extracted. Information about productivity, citations, cluster analysis, keyword co-occurrence, trend topics, and thematic evolution were generated. A total of 448 research articles were included in this analysis. A burst of scholarly activity in the field was noted by the year 2005, peaking in 2017, followed by a remarkable decline to date. Research in the field was hallmarked by consistent and impactful international collaboration, with the US leading in terms of most prolific country, institutions, and total link strength. Thematic evolution in the field points in the direction of more specialized studies on applied pharmacokinetics of available and novel TKIs, particularly for the treatment of lung and breast cancers. Our results delineate a significant advancement in the field of PGx in precision oncology. Notwithstanding the practical challenges to these applications at the point of care, further research, standardization, infrastructure development, and informed policymaking are urgently needed to ensure widespread adoption of PGx.

Machine learning in onco-pharmacogenomics: a path to precision medicine with many challenges

Over the past two decades, Next-Generation Sequencing (NGS) has revolutionized the approach to cancer research. Applications of NGS include the identification of tumor specific alterations that can influence tumor pathobiology and also impact diagnosis, prognosis and therapeutic options. Pharmacogenomics (PGx) studies the role of inheritance of individual genetic patterns in drug response and has taken advantage of NGS technology as it provides access to high-throughput data that can, however, be difficult to manage. Machine learning (ML) has recently been used in the life sciences to discover hidden patterns from complex NGS data and to solve various PGx problems. In this review, we provide a comprehensive overview of the NGS approaches that can be employed and the different PGx studies implicating the use of NGS data. We also provide an excursus of the ML algorithms that can exert a role as fundamental strategies in the PGx field to improve personalized medicine in cancer.

Predicting anticancer synergistic drug combinations based on multi-task learning

BACKGROUND

The discovery of anticancer drug combinations is a crucial work of anticancer treatment. In recent years, pre-screening drug combinations with synergistic effects in a large-scale search space adopting computational methods, especially deep learning methods, is increasingly popular with researchers. Although achievements have been made to predict anticancer synergistic drug combinations based on deep learning, the application of multi-task learning in this field is relatively rare. The successful practice of multi-task learning in various fields shows that it can effectively learn multiple tasks jointly and improve the performance of all the tasks.

METHODS

In this paper, we propose MTLSynergy which is based on multi-task learning and deep neural networks to predict synergistic anticancer drug combinations. It simultaneously learns two crucial prediction tasks in anticancer treatment, which are synergy prediction of drug combinations and sensitivity prediction of monotherapy. And MTLSynergy integrates the classification and regression of prediction tasks into the same model. Moreover, autoencoders are employed to reduce the dimensions of input features.

RESULTS

Compared with the previous methods listed in this paper, MTLSynergy achieves the lowest mean square error of 216.47 and the highest Pearson correlation coefficient of 0.76 on the drug synergy prediction task. On the corresponding classification task, the area under the receiver operator characteristics curve and the area under the precision-recall curve are 0.90 and 0.62, respectively, which are equivalent to the comparison methods. Through the ablation study, we verify that multi-task learning and autoencoder both have a positive effect on prediction performance. In addition, the prediction results of MTLSynergy in many cases are also consistent with previous studies.

CONCLUSION

Our study suggests that multi-task learning is significantly beneficial for both drug synergy prediction and monotherapy sensitivity prediction when combining these two tasks into one model. The ability of MTLSynergy to discover new anticancer synergistic drug combinations noteworthily outperforms other state-of-the-art methods. MTLSynergy promises to be a powerful tool to pre-screen anticancer synergistic drug combinations.

A Transcriptome-Based Precision Oncology Platform for Patient-Therapy Alignment in a Diverse Set of Treatment-Resistant Malignancies

Predicting in vivo response to antineoplastics remains an elusive challenge. We performed a first-of-kind evaluation of two transcriptome-based precision cancer medicine methodologies to predict tumor sensitivity to a comprehensive repertoire of clinically relevant oncology drugs, whose mechanism of action we experimentally assessed in cognate cell lines. We enrolled patients with histologically distinct, poor-prognosis malignancies who had progressed on multiple therapies, and developed low-passage, patient-derived xenograft models that were used to validate 35 patient-specific drug predictions. Both OncoTarget, which identifies high-affinity inhibitors of individual master regulator (MR) proteins, and OncoTreat, which identifies drugs that invert the transcriptional activity of hyperconnected MR modules, produced highly significant 30-day disease control rates (68% and 91%, respectively). Moreover, of 18 OncoTreat-predicted drugs, 15 induced the predicted MR-module activity inversion in vivo. Predicted drugs significantly outperformed antineoplastic drugs selected as unpredicted controls, suggesting these methods may substantively complement existing precision cancer medicine approaches, as also illustrated by a case study.

SIGNIFICANCE

Complementary precision cancer medicine paradigms are needed to broaden the clinical benefit realized through genetic profiling and immunotherapy. In this first-in-class application, we introduce two transcriptome-based tumor-agnostic systems biology tools to predict drug response in vivo. OncoTarget and OncoTreat are scalable for the design of basket and umbrella clinical trials. This article is highlighted in the In This Issue feature, p. 1275.

Benchmarking of protein interaction databases for integration with manually reconstructed signalling network models

Protein interaction databases are critical resources for network bioinformatics and integrating molecular experimental data. Interaction databases may also enable construction of predictive computational models of biological networks, although their fidelity for this purpose is not clear. Here, we benchmark protein interaction databases X2K, Reactome, Pathway Commons, Omnipath and Signor for their ability to recover manually curated edges from three logic-based network models of cardiac hypertrophy, mechano-signalling and fibrosis. Pathway Commons performed best at recovering interactions from manually reconstructed hypertrophy (137 of 193 interactions, 71%), mechano-signalling (85 of 125 interactions, 68%) and fibroblast networks (98 of 142 interactions, 69%). While protein interaction databases successfully recovered central, well-conserved pathways, they performed worse at recovering tissue-specific and transcriptional regulation. This highlights a knowledge gap where manual curation is critical. Finally, we tested the ability of Signor and Pathway Commons to identify new edges that improve model predictions, revealing important roles of protein kinase C autophosphorylation and Ca2+ /calmodulin-dependent protein kinase II phosphorylation of CREB in cardiomyocyte hypertrophy. This study provides a platform for benchmarking protein interaction databases for their utility in network model construction, as well as providing new insights into cardiac hypertrophy signalling. KEY POINTS: Protein interaction databases are used to recover signalling interactions from previously developed network models. The five protein interaction databases benchmarked recovered well-conserved pathways, but did poorly at recovering tissue-specific pathways and transcriptional regulation, indicating the importance of manual curation. We identify new signalling interactions not previously used in the network models, including a role for Ca2+ /calmodulin-dependent protein kinase II phosphorylation of CREB in cardiomyocyte hypertrophy.

OncoLoop: A Network-Based Precision Cancer Medicine Framework

Prioritizing treatments for individual patients with cancer remains challenging, and performing coclinical studies using patient-derived models in real time is often unfeasible. To circumvent these challenges, we introduce OncoLoop, a precision medicine framework that predicts drug sensitivity in human tumors and their preexisting high-fidelity (cognate) model(s) by leveraging drug perturbation profiles. As a proof of concept, we applied OncoLoop to prostate cancer using genetically engineered mouse models (GEMM) that recapitulate a broad spectrum of disease states, including castration-resistant, metastatic, and neuroendocrine prostate cancer. Interrogation of human prostate cancer cohorts by Master Regulator (MR) conservation analysis revealed that most patients with advanced prostate cancer were represented by at least one cognate GEMM-derived tumor (GEMM-DT). Drugs predicted to invert MR activity in patients and their cognate GEMM-DTs were successfully validated in allograft, syngeneic, and patient-derived xenograft (PDX) models of tumors and metastasis. Furthermore, OncoLoop-predicted drugs enhanced the efficacy of clinically relevant drugs, namely, the PD-1 inhibitor nivolumab and the AR inhibitor enzalutamide.

SIGNIFICANCE

OncoLoop is a transcriptomic-based experimental and computational framework that can support rapid-turnaround coclinical studies to identify and validate drugs for individual patients, which can then be readily adapted to clinical practice. This framework should be applicable in many cancer contexts for which appropriate models and drug perturbation data are available. This article is highlighted in the In This Issue feature, p. 247.

Advancing Targeted Protein Degradation via Multiomics Profiling and Artificial Intelligence

Only around 20% of the human proteome is considered to be druggable with small-molecule antagonists. This leaves some of the most compelling therapeutic targets outside the reach of ligand discovery. The concept of targeted protein degradation (TPD) promises to overcome some of these limitations. In brief, TPD is dependent on small molecules that induce the proximity between a protein of interest (POI) and an E3 ubiquitin ligase, causing ubiquitination and degradation of the POI. In this perspective, we want to reflect on current challenges in the field, and discuss how advances in multiomics profiling, artificial intelligence, and machine learning (AI/ML) will be vital in overcoming them. The presented roadmap is discussed in the context of small-molecule degraders but is equally applicable for other emerging proximity-inducing modalities.

Integrating and formatting biomedical data as pre-calculated knowledge graph embeddings in the Bioteque

Biomedical data is accumulating at a fast pace and integrating it into a unified framework is a major challenge, so that multiple views of a given biological event can be considered simultaneously. Here we present the Bioteque, a resource of unprecedented size and scope that contains pre-calculated biomedical descriptors derived from a gigantic knowledge graph, displaying more than 450 thousand biological entities and 30 million relationships between them. The Bioteque integrates, harmonizes, and formats data collected from over 150 data sources, including 12 biological entities (e.g., genes, diseases, drugs) linked by 67 types of associations (e.g., ‘drug treats disease’, ‘gene interacts with gene’). We show how Bioteque descriptors facilitate the assessment of high-throughput protein-protein interactome data, the prediction of drug response and new repurposing opportunities, and demonstrate that they can be used off-the-shelf in downstream machine learning tasks without loss of performance with respect to using original data. The Bioteque thus offers a thoroughly processed, tractable, and highly optimized assembly of the biomedical knowledge available in the public domain.

Biomedical data is accumulating at a fast pace and integrating it into a unified framework is a major challenge. Here, the authors present a resource that contains pre-calculated biomedical descriptors derived from a very large knowledge graph.

A Patient-to-Model-to-Patient (PMP) Cancer Drug Discovery Protocol for Identifying and Validating Therapeutic Agents Targeting Tumor Regulatory Architecture

The current Achilles heel of cancer drug discovery is the inability to forge precise and predictive connections among mechanistic drivers of the cancer cell state, therapeutically significant molecular targets, effective drugs, and responsive patient subgroups. Although advances in molecular biology have helped identify molecular markers and stratify patients into molecular subtypes, these associational strategies typically fail to provide a mechanistic rationale to identify cancer vulnerabilities. Recently, integrative systems biology methodologies have been used to reverse engineer cellular networks and identify master regulators (MRs), proteins whose activity is both necessary and sufficient to implement phenotypic states under physiological and pathological conditions, which are organized into highly interconnected regulatory modules called tumor checkpoints. Because of their functional relevance, MRs represent ideal pharmacological targets and biomarkers. Here, we present a six-step patient-to-model-to-patient protocol that employs computational and experimental methodologies to reconstruct and interrogate the regulatory logic of human cancer cells for identifying and therapeutically targeting the tumor checkpoint with novel as well as existing pharmacological agents. This protocol systematically identifies, from specific patient tumor samples, the MRs that comprise the tumor checkpoint. Then, it identifies in vitro and in vivo models that, by recapitulating the patient's tumor checkpoint, constitute the appropriate cell lines and xenografts to further elucidate the tissue context-specific drug mechanism of action (MOA) and permit precise, biomarker-based preclinical validations of drug efficacy. The combination of determination of a drug's context-specific MOA and precise identification of patients' tumor checkpoints provides a personalized, mechanism-based biomarker to enrich prospective clinical trials with patients likely to respond. © 2022 The Authors. Current Protocols published by Wiley Periodicals LLC.

A model for network-based identification and pharmacological targeting of aberrant, replication-permissive transcriptional programs induced by viral infection

SARS-CoV-2 hijacks the host cell transcriptional machinery to induce a phenotypic state amenable to its replication. Here we show that analysis of Master Regulator proteins representing mechanistic determinants of the gene expression signature induced by SARS-CoV-2 in infected cells revealed coordinated inactivation of Master Regulators enriched in physical interactions with SARS-CoV-2 proteins, suggesting their mechanistic role in maintaining a host cell state refractory to virus replication. To test their functional relevance, we measured SARS-CoV-2 replication in epithelial cells treated with drugs predicted to activate the entire repertoire of repressed Master Regulators, based on their experimentally elucidated, context-specific mechanism of action. Overall, 15 of the 18 drugs predicted to be effective by this methodology induced significant reduction of SARS-CoV-2 replication, without affecting cell viability. This model for host-directed pharmacological therapy is fully generalizable and can be deployed to identify drugs targeting host cell-based Master Regulator signatures induced by virtually any pathogen.

Considerations and challenges for sex-aware drug repurposing

Sex differences are essential factors in disease etiology and manifestation in many diseases such as cardiovascular disease, cancer, and neurodegeneration [33]. The biological influence of sex differences (including genomic, epigenetic, hormonal, immunological, and metabolic differences between males and females) and the lack of biomedical studies considering sex differences in their study design has led to several policies. For example, the National Institute of Health's (NIH) sex as a biological variable (SABV) and Sex and Gender Equity in Research (SAGER) policies to motivate researchers to consider sex differences [204]. However, drug repurposing, a promising alternative to traditional drug discovery by identifying novel uses for FDA-approved drugs, lacks sex-aware methods that can improve the identification of drugs that have sex-specific responses [7, 11, 14, 33]. Sex-aware drug repurposing methods either select drug candidates that are more efficacious in one sex or deprioritize drug candidates based on if they are predicted to cause a sex-bias adverse event (SBAE), unintended therapeutic effects that are more likely to occur in one sex. Computational drug repurposing methods are encouraging approaches to develop for sex-aware drug repurposing because they can prioritize sex-specific drug candidates or SBAEs at lower cost and time than traditional drug discovery. Sex-aware methods currently exist for clinical, genomic, and transcriptomic information [1, 7, 155]. They have not expanded to other data types, such as DNA variation, which has been beneficial in other drug repurposing methods that do not consider sex [114]. Additionally, some sex-aware methods suffer from poorer performance because a disproportionate number of male and female samples are available to train computational methods [7]. However, there is development potential for several different categories (i.e., data mining, ligand binding predictions, molecular associations, and networks). Low-dimensional representations of molecular association and network approaches are also especially promising candidates for future sex-aware drug repurposing methodologies because they reduce the multiple hypothesis testing burden and capture sex-specific variation better than the other methods [151, 159]. Here we review how sex influences drug response, the current state of drug repurposing including with respect to sex-bias drug response, and how model organism study design choices influence drug repurposing validation.