Systematic interrogation of diverse Omic data reveals interpretable, robust, and generalizable transcriptomic features of clinically successful therapeutic targets

Navigating the complexities of drug development for inflammatory bowel disease

Inflammatory bowel disease (IBD) - consisting of ulcerative colitis and Crohn's disease - is a complex, heterogeneous, immune-mediated inflammatory condition with a multifactorial aetiopathogenesis. Despite therapeutic advances in this arena, a ceiling effect has been reached with both single-agent monoclonal antibodies and advanced small molecules. Therefore, there is a need to identify novel targets, and the development of companion biomarkers to select responders is vital. In this Perspective, we examine how advances in machine learning and tissue engineering could be used at the preclinical stage where attrition rates are high. For novel agents reaching clinical trials, we explore factors decelerating progression, particularly the decline in IBD trial recruitment, and assess how innovative approaches such as reconfiguring trial designs, harmonizing end points and incorporating digital technologies into clinical trials can address this. Harnessing opportunities at each stage of the drug development process may allow for incremental gains towards more effective therapies.

A method to improve the prediction performance of cancer-gene association by screening negative training samples through gene network data

This work focuses on data sampling in cancer-gene association prediction. Currently, researchers are using machine learning methods to predict genes that are more likely to produce cancer-causing mutations. To improve the performance of machine learning models, methods have been proposed, one of which is to improve the quality of the training data. Existing methods focus mainly on positive data, i.e. cancer driver genes, for screening selection. This paper proposes a low-cancer-related gene screening method based on gene network and graph theory algorithms to improve the negative samples selection. Genetic data with low cancer correlation is used as negative training samples. After experimental verification, using the negative samples screened by this method to train the cancer gene classification model can improve prediction performance. The biggest advantage of this method is that it can be easily combined with other methods that focus on enhancing the quality of positive training samples. It has been demonstrated that significant improvement is achieved by combining this method with three state-of-the-arts cancer gene prediction methods.

Long Noncoding RNA VLDLR-AS1 Levels in Serum Correlate with Combat-Related Chronic Mild Traumatic Brain Injury and Depression Symptoms in US Veterans

More than 75% of traumatic brain injuries (TBIs) are mild (mTBI) and military service members often experience repeated combat-related mTBI. The chronic comorbidities concomitant with repetitive mTBI (rmTBI) include depression, post-traumatic stress disorder or neurological dysfunction. This study sought to determine a long noncoding RNA (lncRNA) expression signature in serum samples that correlated with rmTBI years after the incidences. Serum samples were obtained from Long-Term Impact of Military-Relevant Brain-Injury Consortium Chronic Effects of Neurotrauma Consortium (LIMBIC CENC) repository, from participants unexposed to TBI or who had rmTBI. Four lncRNAs were identified as consistently present in all samples, as detected via droplet digital PCR and packaged in exosomes enriched for CNS origin. The results, using qPCR, demonstrated that the lncRNA VLDLR-AS1 levels were significantly lower among individuals with rmTBI compared to those with no lifetime TBI. ROC analysis determined an AUC of 0.74 (95% CI: 0.6124 to 0.8741; p = 0.0012). The optimal cutoff for VLDLR-AS1 was ≤153.8 ng. A secondary analysis of clinical data from LIMBIC CENC was conducted to evaluate the psychological symptom burden, and the results show that lncRNAs VLDLR-AS1 and MALAT1 are correlated with symptoms of depression. In conclusion, lncRNA VLDLR-AS1 may serve as a blood biomarker for identifying chronic rmTBI and depression in patients.

High-Throughput Synthesis and Machine Learning Assisted Design of Photodegradable Hydrogels

Due to the large chemical space, the design of functional and responsive soft materials poses many challenges but also offers a wide range of opportunities in terms of the scope of possible properties. Herein, an experimental workflow for miniaturized combinatorial high-throughput screening of functional hydrogel libraries is reported. The data created from the analysis of the photodegradation process of more than 900 different types of hydrogel pads are used to train a machine learning model for automated decision making. Through iterative model optimization based on Bayesian optimization, a substantial improvement in response properties is achieved and thus expanded the scope of material properties obtainable within the chemical space of hydrogels in the study. It is therefore demonstrated that the potential of combining miniaturized high-throughput experiments with smart optimization algorithms for cost and time efficient optimization of materials properties.

Drug approval prediction based on the discrepancy in gene perturbation effects between cells and humans

BACKGROUND

Poor translation between in vitro and clinical studies due to the cells/humans discrepancy in drug target perturbation effects leads to safety failures in clinical trials, thus increasing drug development costs and reducing patients' life quality. Therefore, developing a predictive model for drug approval considering the cells/humans discrepancy is needed to reduce drug attrition rates in clinical trials.

METHODS

Our machine learning framework predicts drug approval in clinical trials based on the cells/humans discrepancy in drug target perturbation effects. To evaluate the discrepancy to predict drug approval (1404 approved and 1070 unapproved drugs), we analysed CRISPR-Cas9 knockout and loss-of-function mutation rate-based gene perturbation effects on cells and humans, respectively. To validate the risk of drug targets with the cells/humans discrepancy, we examined the targets of failed and withdrawn drugs due to safety problems.

FINDINGS

Drug approvals in clinical trials were correlated with the cells/humans discrepancy in gene perturbation effects. Genes tolerant to perturbation effects on cells but intolerant to those on humans were associated with failed drug targets. Furthermore, genes with the cells/humans discrepancy were related to drugs withdrawn due to severe side effects. Motivated by previous studies assessing drug safety through chemical properties, we improved drug approval prediction by integrating chemical information with the cells/humans discrepancy.

INTERPRETATION

The cells/humans discrepancy in gene perturbation effects facilitates drug approval prediction and explains drug safety failures in clinical trials.

FUNDING

S.K. received grants from the Korean National Research Foundation (2021R1A2B5B01001903 and 2020R1A6A1A03047902) and IITP (2019-0-01906, Artificial Intelligence Graduate School Program, POSTECH).

PINNED: identifying characteristics of druggable human proteins using an interpretable neural network

The identification of human proteins that are amenable to pharmacologic modulation without significant off-target effects remains an important unsolved challenge. Computational methods have been devised to identify features which distinguish between "druggable" and "undruggable" proteins, finding that protein sequence, tissue and cellular localization, biological role, and position in the protein-protein interaction network are all important discriminant factors. However, many prior efforts to automate the assessment of protein druggability suffer from low performance or poor interpretability. We developed a neural network-based machine learning model capable of generating druggability sub-scores based on each of four distinct categories, combining them to form an overall druggability score. The model achieves an excellent performance in separating drugged and undrugged proteins in the human proteome, with an area under the receiver operating characteristic (AUC) of 0.95. Our use of multiple sub-scores allows the assessment of potential protein targets of interest based on distinct contributors to druggability, leading to a more interpretable and holistic model to identify novel targets.

Pharma's Bio-AI revolution

Drug development has become unbearably slow and expensive. A key underlying problem is the clinical prediction challenge: the inability to predict which drug candidates will be safe in the human body and for whom. Recently, a dramatic regulatory change has removed FDA's mandated reliance on antiquated, ineffective animal studies. A new frontier is an integration of several disruptive technologies [machine learning (ML), patient-on-chip, real-time sensing, and stem cells], which when integrated, have the potential to address this challenge, drastically cutting the time and cost of developing drugs, and tailoring them to individual patients.

Molecular Landscape of Tourette's Disorder

Tourette's disorder (TD) is a highly heritable childhood-onset neurodevelopmental disorder and is caused by a complex interplay of multiple genetic and environmental factors. Yet, the molecular mechanisms underlying the disorder remain largely elusive. In this study, we used the available omics data to compile a list of TD candidate genes, and we subsequently conducted tissue/cell type specificity and functional enrichment analyses of this list. Using genomic data, we also investigated genetic sharing between TD and blood and cerebrospinal fluid (CSF) metabolite levels. Lastly, we built a molecular landscape of TD through integrating the results from these analyses with an extensive literature search to identify the interactions between the TD candidate genes/proteins and metabolites. We found evidence for an enriched expression of the TD candidate genes in four brain regions and the pituitary. The functional enrichment analyses implicated two pathways ('cAMP-mediated signaling' and 'Endocannabinoid Neuronal Synapse Pathway') and multiple biological functions related to brain development and synaptic transmission in TD etiology. Furthermore, we found genetic sharing between TD and the blood and CSF levels of 39 metabolites. The landscape of TD not only provides insights into the (altered) molecular processes that underlie the disease but, through the identification of potential drug targets (such as FLT3, NAALAD2, CX3CL1-CX3CR1, OPRM1, and HRH2), it also yields clues for developing novel TD treatments.

Machine Learning in Drug Discovery: A Review

This review provides the feasible literature on drug discovery through ML tools and techniques that are enforced in every phase of drug development to accelerate the research process and deduce the risk and expenditure in clinical trials. Machine learning techniques improve the decision-making in pharmaceutical data across various applications like QSAR analysis, hit discoveries, de novo drug architectures to retrieve accurate outcomes. Target validation, prognostic biomarkers, digital pathology are considered under problem statements in this review. ML challenges must be applicable for the main cause of inadequacy in interpretability outcomes that may restrict the applications in drug discovery. In clinical trials, absolute and methodological data must be generated to tackle many puzzles in validating ML techniques, improving decision-making, promoting awareness in ML approaches, and deducing risk failures in drug discovery.

Translational precision medicine: an industry perspective

In the era of precision medicine, digital technologies and artificial intelligence, drug discovery and development face unprecedented opportunities for product and business model innovation, fundamentally changing the traditional approach of how drugs are discovered, developed and marketed. Critical to this transformation is the adoption of new technologies in the drug development process, catalyzing the transition from serendipity-driven to data-driven medicine. This paradigm shift comes with a need for both translation and precision, leading to a modern Translational Precision Medicine approach to drug discovery and development. Key components of Translational Precision Medicine are multi-omics profiling, digital biomarkers, model-based data integration, artificial intelligence, biomarker-guided trial designs and patient-centric companion diagnostics. In this review, we summarize and critically discuss the potential and challenges of Translational Precision Medicine from a cross-industry perspective.

Artificial intelligence in drug design: algorithms, applications, challenges and ethics

The discovery paradigm of drugs is rapidly growing due to advances in machine learning (ML) and artificial intelligence (AI). This review covers myriad faces of AI and ML in drug design. There is a plethora of AI algorithms, the most common of which are summarized in this review. In addition, AI is fraught with challenges that are highlighted along with plausible solutions to them. Examples are provided to illustrate the use of AI and ML in drug discovery and in predicting drug properties such as binding affinities and interactions, solubility, toxicology, blood–brain barrier permeability and chemical properties. The review also includes examples depicting the implementation of AI and ML in tackling intractable diseases such as COVID-19, cancer and Alzheimer’s disease. Ethical considerations and future perspectives of AI are also covered in this review.

Semi-supervised Hierarchical Drug Embedding in Hyperbolic Space

Learning accurate drug representations is essential for tasks such as computational drug repositioning and prediction of drug side effects. A drug hierarchy is a valuable source that encodes knowledge of relations among drugs in a tree-like structure where drugs that act on the same organs, treat the same disease, or bind to the same biological target are grouped together. However, its utility in learning drug representations has not yet been explored, and currently described drug representations cannot place novel molecules in a drug hierarchy. Here, we develop a semi-supervised drug embedding that incorporates two sources of information: (1) underlying chemical grammar that is inferred from chemical structures of drugs and drug-like molecules (unsupervised) and (2) hierarchical relations that are encoded in an expert-crafted hierarchy of approved drugs (supervised). We use the Variational Auto-Encoder (VAE) framework to encode the chemical structures of molecules and use the drug-drug similarity information obtained from the hierarchy to induce the clustering of drugs in hyperbolic space. The hyperbolic space is amenable for encoding hierarchical relations. Both quantitative and qualitative results support that the learned drug embedding can accurately reproduce the chemical structure and recapitulate the hierarchical relations among drugs. Furthermore, our approach can infer the pharmacological properties of novel molecules by retrieving similar drugs from the embedding space. We demonstrate that our drug embedding can predict new uses and discover new side effects of existing drugs. We show that it significantly outperforms comparison methods in both tasks.

In silico analysis of alternative splicing on drug-target gene interactions

Identifying and evaluating the right target are the most important factors in early drug discovery phase. Most studies focus on one protein ignoring the multiple splice-variant or protein-isoforms, which might contribute to unexpected therapeutic activity or adverse side effects. Here, we present computational analysis of cancer drug-target interactions affected by alternative splicing. By integrating information from publicly available databases, we curated 883 FDA approved or investigational stage small molecule cancer drugs that target 1,434 different genes, with an average of 5.22 protein isoforms per gene. Of these, 618 genes have ≥5 annotated protein-isoforms. By analyzing the interactions with binding pocket information, we found that 76% of drugs either miss a potential target isoform or target other isoforms with varied expression in multiple normal tissues. We present sequence and structure level alignments at isoform-level and make this information publicly available for all the curated drugs. Structure-level analysis showed ligand binding pocket architectures differences in size, shape and electrostatic parameters between isoforms. Our results emphasize how potentially important isoform-level interactions could be missed by solely focusing on the canonical isoform, and suggest that on- and off-target effects at isoform-level should be investigated to enhance the productivity of drug-discovery research.

Applications of machine learning in drug discovery and development

Drug discovery and development pipelines are long, complex and depend on numerous factors. Machine learning (ML) approaches provide a set of tools that can improve discovery and decision making for well-specified questions with abundant, high-quality data. Opportunities to apply ML occur in all stages of drug discovery. Examples include target validation, identification of prognostic biomarkers and analysis of digital pathology data in clinical trials. Applications have ranged in context and methodology, with some approaches yielding accurate predictions and insights. The challenges of applying ML lie primarily with the lack of interpretability and repeatability of ML-generated results, which may limit their application. In all areas, systematic and comprehensive high-dimensional data still need to be generated. With ongoing efforts to tackle these issues, as well as increasing awareness of the factors needed to validate ML approaches, the application of ML can promote data-driven decision making and has the potential to speed up the process and reduce failure rates in drug discovery and development.

Tissue-specific genes as an underutilized resource in drug discovery

Tissue-specific genes are believed to be good drug targets due to improved safety. Here we show that this intuitive notion is not reflected in phase 1 and 2 clinical trials, despite the historic success of tissue-specific targets and their 2.3-fold overrepresentation among targets of marketed non-oncology drugs. We compare properties of tissue-specific genes and drug targets. We show that tissue-specificity of the target may also be related to efficacy of the drug. The relationship may be indirect (enrichment in Mendelian disease and PTVesc genes) or direct (elevated betweenness centrality scores for tissue-specifically produced enzymes and secreted proteins). Reduced evolutionary conservation of tissue-specific genes may represent a bottleneck for drug projects, prompting development of novel models with smaller evolutionary gap to humans. We show that the opportunities to identify tissue-specific drug targets are not exhausted and discuss potential use cases for tissue-specific genes in drug research.

Predicting clinically promising therapeutic hypotheses using tensor factorization

BACKGROUND

Determining which target to pursue is a challenging and error-prone first step in developing a therapeutic treatment for a disease, where missteps are potentially very costly given the long-time frames and high expenses of drug development. With current informatics technology and machine learning algorithms, it is now possible to computationally discover therapeutic hypotheses by predicting clinically promising drug targets based on the evidence associating drug targets with disease indications. We have collected this evidence from Open Targets and additional databases that covers 17 sources of evidence for target-indication association and represented the data as a tensor of 21,437 × 2211 × 17.

RESULTS

As a proof-of-concept, we identified examples of successes and failures of target-indication pairs in clinical trials across 875 targets and 574 disease indications to build a gold-standard data set of 6140 known clinical outcomes. We designed and executed three benchmarking strategies to examine the performance of multiple machine learning models: Logistic Regression, LASSO, Random Forest, Tensor Factorization and Gradient Boosting Machine. With 10-fold cross-validation, tensor factorization achieved AUROC = 0.82 ± 0.02 and AUPRC = 0.71 ± 0.03. Across multiple validation schemes, this was comparable or better than other methods.

CONCLUSION

In this work, we benchmarked a machine learning technique called tensor factorization for the problem of predicting clinical outcomes of therapeutic hypotheses. Results have shown that this method can achieve equal or better prediction performance compared with a variety of baseline models. We demonstrate one application of the method to predict outcomes of trials on novel indications of approved drug targets. This work can be expanded to targets and indications that have never been clinically tested and proposing novel target-indication hypotheses. Our proposed biologically-motivated cross-validation schemes provide insight into the robustness of the prediction performance. This has significant implications for all future methods that try to address this seminal problem in drug discovery.