Machine Learning Informs RNA-Binding Chemical Space

RNATACs: Multispecific small molecules targeting RNA by induced proximity

RNA-targeting small molecules (rSMs) have become an attractive modality to tackle traditionally undruggable proteins and expand the druggable space. Among many innovative concepts, RNA-targeting chimeras (RNATACs) represent a new class of multispecific, induced proximity small molecules that act by chemically bringing RNA targets into proximity with an endogenous RNA effector, such as a ribonuclease (RNase). Depending on the RNA effector, RNATACs can alter the stability, localization, translation, or splicing of the target RNA. Although still in its infancy, this new modality has the potential for broad applications in the future to treat diseases with high unmet need. In this review, we discuss potential advantages of RNATACs, recent progress in the field, and challenges to this cutting-edge technology.

Competitive Microarray Screening Reveals Functional Ligands for the DHX15 RNA G-Quadruplex

RNAs are increasingly considered valuable therapeutic targets, and the development of methods to identify and validate both RNA targets and ligands is more important than ever. Here, we utilized a bioinformatic approach to identify a hairpin-containing RNA G-quadruplex (rG4) in the 5' untranslated region (5' UTR) of DHX15 mRNA. By using a novel competitive small molecule microarray (SMM) approach, we identified a compound that specifically binds to the DHX15 rG4 (K D = 12.6 ± 1.0 μM). This rG4 directly impacts translation of a DHX15 reporter mRNA in vitro, and binding of our compound (F1) to the structure inhibits translation up to 57% (IC50 = 22.9 ± 3.8 μM). This methodology allowed us to identify and target the mRNA of a cancer-relevant helicase with no known inhibitors. Our target identification method and the novelty of our screening approach make our work informative for future development of novel small molecule cancer therapeutics for RNA targets.

Advances in machine-learning approaches to RNA-targeted drug design

RNA molecules play multifaceted functional and regulatory roles within cells and have garnered significant attention in recent years as promising therapeutic targets. With remarkable successes achieved by artificial intelligence (AI) in different fields such as computer vision and natural language processing, there is a growing imperative to harness AI's potential in computer-aided drug design (CADD) to discover novel drug compounds that target RNA. Although machine-learning (ML) approaches have been widely adopted in the discovery of small molecules targeting proteins, the application of ML approaches to model interactions between RNA and small molecule is still in its infancy. Compared to protein-targeted drug discovery, the major challenges in ML-based RNA-targeted drug discovery stem from the scarcity of available data resources. With the growing interest and the development of curated databases focusing on interactions between RNA and small molecule, the field anticipates a rapid growth and the opening of a new avenue for disease treatment. In this review, we aim to provide an overview of recent advancements in computationally modeling RNA-small molecule interactions within the context of RNA-targeted drug discovery, with a particular emphasis on methodologies employing ML techniques.

A QSAR study for predicting malformation in zebrafish embryo

BACKGROUND

Developmental toxicity tests are extremely expensive, require a large number of animals, and are time-consuming. It is necessary to develop a new approach to simplify the analysis of developmental endpoints. One of these endpoints is malformation, and one group of ongoing methods for simplifying is in silico models. In this study, we aim to develop a quantitative structure-activity relationship (QSAR) model and identify the best algorithm for predicting malformations, as well as the most important and effective physicochemical properties associated with malformation.

METHODS

The dataset was extracted from a reliable database called COMPTOX. Physicochemical properties (descriptors) were calculated using Mordred and RDKit chemoinformatics software. The data were cleaned, preprocessed, and then split into training and testing sets. Machine learning algorithms, such as gradient boosting model (GBM) and logistic regression (LR), as well as deep learning models, including multilayer perceptron (MLP) and neural networks (NNs) trained with train set data and different sets of descriptors. The models were then validated with test set and various statistical parameters, such as Matthew's correlation coefficient (MCC) and balanced accuracy (BAC) score, were used to compare the models.

RESULTS

A set of descriptors containing with 78% AUC was identified as the best set of descriptors. Gradient boosting was determined to be the best algorithm with 78% predictive power.

CONCLUSIONS

The descriptors that were the most effective for developing models directly impact the mechanism of malformation, and GBM is the best model due to its MCC and BAC.

Recent applications of artificial intelligence in RNA-targeted small molecule drug discovery

INTRODUCTION

Targeting RNAs with small molecules offers an alternative to the conventional protein-targeted drug discovery and can potentially address unmet and emerging medical needs. The recent rise of interest in the strategy has already resulted in large amounts of data on disease associated RNAs, as well as on small molecules that bind to such RNAs. Artificial intelligence (AI) approaches, including machine learning and deep learning, present an opportunity to speed up the discovery of RNA-targeted small molecules by improving decision-making efficiency and quality.

AREAS COVERED

The topics described in this review include the recent applications of AI in the identification of RNA targets, RNA structure determination, screening of chemical compound libraries, and hit-to-lead optimization. The impact and limitations of the recent AI applications are discussed, along with an outlook on the possible applications of next-generation AI tools for the discovery of novel RNA-targeted small molecule drugs.

EXPERT OPINION

Key areas for improvement include developing AI tools for understanding RNA dynamics and RNA - small molecule interactions. High-quality and comprehensive data still need to be generated especially on the biological activity of small molecules that target RNAs.

LncRNAs: the good, the bad, and the unknown

Long non-coding RNAs (lncRNAs) are significant contributors in maintaining genomic integrity through epigenetic regulation. LncRNAs can interact with chromatin-modifying complexes in both cis and trans pathways, drawing them to specific genomic loci and influencing gene expression via DNA methylation, histone modifications, and chromatin remodeling. They can also operate as building blocks to assemble different chromatin-modifying components, facilitating their interactions and gene regulatory functions. Deregulation of these molecules has been associated with various human diseases, including cancer, cardiovascular disease, and neurological disorders. Thus, lncRNAs are implicated as potential diagnostic indicators and therapeutic targets. This review discusses the current understanding of how lncRNAs mediate epigenetic control, genomic integrity, and their putative functions in disease pathogenesis.

Reliable method for predicting the binding affinity of RNA-small molecule interactions using machine learning

Ribonucleic acids (RNAs) play important roles in cellular regulation. Consequently, dysregulation of both coding and non-coding RNAs has been implicated in several disease conditions in the human body. In this regard, a growing interest has been observed to probe into the potential of RNAs to act as drug targets in disease conditions. To accelerate this search for disease-associated novel RNA targets and their small molecular inhibitors, machine learning models for binding affinity prediction were developed specific to six RNA subtypes namely, aptamers, miRNAs, repeats, ribosomal RNAs, riboswitches and viral RNAs. We found that differences in RNA sequence composition, flexibility and polar nature of RNA-binding ligands are important for predicting the binding affinity. Our method showed an average Pearson correlation (r) of 0.83 and a mean absolute error of 0.66 upon evaluation using the jack-knife test, indicating their reliability despite the low amount of data available for several RNA subtypes. Further, the models were validated with external blind test datasets, which outperform other existing quantitative structure-activity relationship (QSAR) models. We have developed a web server to host the models, RNA-Small molecule binding Affinity Predictor, which is freely available at: https://web.iitm.ac.in/bioinfo2/RSAPred/.

Functional microRNA-targeting drug discovery by graph-based deep learning

MicroRNAs are recognized as key drivers in many cancers but targeting them with small molecules remains a challenge. We present RiboStrike, a deep-learning framework that identifies small molecules against specific microRNAs. To demonstrate its capabilities, we applied it to microRNA-21 (miR-21), a known driver of breast cancer. To ensure selectivity toward miR-21, we performed counter-screens against miR-122 and DICER. Auxiliary models were used to evaluate toxicity and rank the candidates. Learning from various datasets, we screened a pool of nine million molecules and identified eight, three of which showed anti-miR-21 activity in both reporter assays and RNA sequencing experiments. Target selectivity of these compounds was assessed using microRNA profiling and RNA sequencing analysis. The top candidate was tested in a xenograft mouse model of breast cancer metastasis, demonstrating a significant reduction in lung metastases. These results demonstrate RiboStrike's ability to nominate compounds that target the activity of miRNAs in cancer.

Reactivity-based RNA profiling for analyzing transcriptome interactions of small molecules in human cells

Protein-targeted small-molecule drugs may unintentionally bind intracellular RNA, contributing to drug toxicity. Moreover, new drugs are actively sought for intentionally targeting RNA. Here, we present a protocol to globally profile RNA-drug interactions in human cells using acylating probes and next-generation sequencing. We describe steps for cell culture, target acylation, library preparation, and sequencing. Detailed bioinformatic analyses identify drug-binding RNA loci in ∼16,000 poly(A)+ human transcripts. This streamlined workflow identifies RNA-drug interactions at single-nucleotide resolution, revealing widespread transcriptome interactions of drugs. For complete details on the use and execution of this protocol, please refer to Fang et al.1.

Pervasive transcriptome interactions of protein-targeted drugs

The off-target toxicity of drugs targeted to proteins imparts substantial health and economic costs. Proteome interaction studies can reveal off-target effects with unintended proteins; however, little attention has been paid to intracellular RNAs as potential off-targets that may contribute to toxicity. To begin to assess this, we developed a reactivity-based RNA profiling methodology and applied it to uncover transcriptome interactions of a set of Food and Drug Administration-approved small-molecule drugs in vivo. We show that these protein-targeted drugs pervasively interact with the human transcriptome and can exert unintended biological effects on RNA functions. In addition, we show that many off-target interactions occur at RNA loci associated with protein binding and structural changes, allowing us to generate hypotheses to infer the biological consequences of RNA off-target binding. The results suggest that rigorous characterization of drugs' transcriptome interactions may help assess target specificity and potentially avoid toxicity and clinical failures.

Nanoporous Crystalline Materials for the Recognition and Applications of Nucleic Acids

Nucleic acid plays a crucial role in countless biological processes. Hence, there is great interest in its detection and analysis in various fields from chemistry, biology, to medicine. Nanoporous crystalline materials exhibit enormous potential as an effective platform for nucleic acid recognition and application. These materials have highly ordered and uniform pore structures, as well as adjustable surface chemistry and pore size, making them good carriers for nucleic acid extraction, detection, and delivery. In this review, the latest developments in nanoporous crystalline materials, including metal organic frameworks (MOFs), covalent organic frameworks (COFs), and supramolecular organic frameworks (SOFs) for nucleic acid recognition and applications are discussed. Different strategies for functionalizing these materials are explored to specifically identify nucleic acid targets. Their applications in selective separation and detection of nucleic acids are highlighted. They can also be used as DNA/RNA sensors, gene delivery agents, host DNAzymes, and in DNA-based computing. Other applications include catalysis, data storage, and biomimetics. The development of novel nanoporous crystalline materials with enhanced biocompatibility has opened up new avenues in the fields of nucleic acid analysis and therapy, paving the way for the development of sensitive, selective, and cost-effective diagnostic and therapeutic tools with widespread applications.

Probing Bioactive Chemical Space to Discover RNA-Targeted Small Molecules

Small molecules have become increasingly recognized as invaluable tools to study RNA structure and function and to develop RNA-targeted therapeutics. To rationally design RNA-targeting ligands, a comprehensive understanding and explicit testing of small molecule properties that govern molecular recognition is crucial. To date, most studies have primarily evaluated properties of small molecules that bind RNA in vitro, with little to no assessment of properties that are distinct to selective and bioactive RNA-targeted ligands. Therefore, we curated an RNA-focused library, termed the Duke RNA-Targeted Library (DRTL), that was biased towards the physicochemical and structural properties of biologically active and non-ribosomal RNA-targeted small molecules. The DRTL represents one of the largest academic RNA-focused small molecule libraries curated to date with more than 800 small molecules. These ligands were selected using computational approaches that measure similarity to known bioactive RNA ligands and that diversify the molecules within this space. We evaluated DRTL binding in vitro to a panel of four RNAs using two optimized fluorescent indicator displacement assays, and we successfully identified multiple small molecule hits, including several novel scaffolds for RNA. The DRTL has and will continue to provide insights into biologically relevant RNA chemical space, such as the identification of additional RNA-privileged scaffolds and validation of RNA-privileged molecular features. Future DRTL screening will focus on expanding both the targets and assays used, and we welcome collaboration from the scientific community. We envision that the DRTL will be a valuable resource for the discovery of RNA-targeted chemical probes and therapeutic leads.

Recent trends in RNA informatics: a review of machine learning and deep learning for RNA secondary structure prediction and RNA drug discovery

Computational analysis of RNA sequences constitutes a crucial step in the field of RNA biology. As in other domains of the life sciences, the incorporation of artificial intelligence and machine learning techniques into RNA sequence analysis has gained significant traction in recent years. Historically, thermodynamics-based methods were widely employed for the prediction of RNA secondary structures; however, machine learning-based approaches have demonstrated remarkable advancements in recent years, enabling more accurate predictions. Consequently, the precision of sequence analysis pertaining to RNA secondary structures, such as RNA-protein interactions, has also been enhanced, making a substantial contribution to the field of RNA biology. Additionally, artificial intelligence and machine learning are also introducing technical innovations in the analysis of RNA-small molecule interactions for RNA-targeted drug discovery and in the design of RNA aptamers, where RNA serves as its own ligand. This review will highlight recent trends in the prediction of RNA secondary structure, RNA aptamers and RNA drug discovery using machine learning, deep learning and related technologies, and will also discuss potential future avenues in the field of RNA informatics.

Discovery of Small Molecules Targeting the Frameshifting Element RNA in SARS-CoV-2 Viral Genome

Targeting structured RNA elements in the SARS-CoV-2 viral genome with small molecules is an attractive strategy for pharmacological control over viral replication. In this work, we report the discovery of small molecules that target the frameshifting element (FSE) in the SARS-CoV-2 RNA genome using high-throughput small-molecule microarray (SMM) screening. A new class of aminoquinazoline ligands for the SARS-CoV-2 FSE are synthesized and characterized using multiple orthogonal biophysical assays and structure-activity relationship (SAR) studies. This work reveals compounds with mid-micromolar binding affinity (KD = 60 ± 6 μM) to the FSE RNA and supports a binding mode distinct from previously reported FSE binders MTDB and merafloxacin. In addition, compounds are active in in vitro dual-luciferase and in-cell dual-fluorescent-reporter frameshifting assays, highlighting the promise of targeting structured elements of RNAs with druglike compounds to alter expression of viral proteins.

Proximity-Induced Nucleic Acid Degrader (PINAD) Approach to Targeted RNA Degradation Using Small Molecules

Nature has evolved intricate machinery to target and degrade RNA, and some of these molecular mechanisms can be adapted for therapeutic use. Small interfering RNAs and RNase H-inducing oligonucleotides have yielded therapeutic agents against diseases that cannot be tackled using protein-centered approaches. Because these therapeutic agents are nucleic acid-based, they have several inherent drawbacks which include poor cellular uptake and stability. Here we report a new approach to target and degrade RNA using small molecules, proximity-induced nucleic acid degrader (PINAD). We have utilized this strategy to design two families of RNA degraders which target two different RNA structures within the genome of SARS-CoV-2: G-quadruplexes and the betacoronaviral pseudoknot. We demonstrate that these novel molecules degrade their targets using in vitro, in cellulo, and in vivo SARS-CoV-2 infection models. Our strategy allows any RNA binding small molecule to be converted into a degrader, empowering RNA binders that are not potent enough to exert a phenotypic effect on their own. PINAD raises the possibility of targeting and destroying any disease-related RNA species, which can greatly expand the space of druggable targets and diseases.

Recognizing the power of machine learning and other computational methods to accelerate progress in small molecule targeting of RNA

RNA structures regulate a wide range of processes in biology and disease, yet small molecule chemical probes or drugs that can modulate these functions are rare. Machine learning and other computational methods are well poised to fill gaps in knowledge and overcome the inherent challenges in RNA targeting, such as the dynamic nature of RNA and the difficulty of obtaining RNA high-resolution structures. Successful tools to date include principal component analysis, linear discriminate analysis, k-nearest neighbor, artificial neural networks, multiple linear regression, and many others. Employment of these tools has revealed critical factors for selective recognition in RNA:small molecule complexes, predictable differences in RNA- and protein-binding ligands, and quantitative structure activity relationships that allow the rational design of small molecules for a given RNA target. Herein we present our perspective on the value of using machine learning and other computation methods to advance RNA:small molecule targeting, including select examples and their validation as well as necessary and promising future directions that will be key to accelerate discoveries in this important field.

Contemporary Progress and Opportunities in RNA-Targeted Drug Discovery

The surprising discovery that RNAs are the predominant gene products to emerge from the human genome catalyzed a renaissance in RNA biology. It is now well-understood that RNAs act as more than just a messenger and comprise a large and diverse family of ribonucleic acids of differing sizes, structures, and functions. RNAs play expansive roles in the cell, contributing to the regulation and fine-tuning of nearly all aspects of gene expression and genome architecture. In line with the significance of these functions, we have witnessed an explosion in discoveries connecting RNAs with a variety of human diseases. Consequently, the targeting of RNAs, and more broadly RNA biology, has emerged as an untapped area of drug discovery, making the search for RNA-targeted therapeutics of great interest. In this Microperspective, I highlight contemporary learnings in the field and present my views on how to catapult us toward the systematic discovery of RNA-targeted medicines.