Target prediction

RNA is crucial for the regulation of numerous cellular processes and functions. With the in-depth study of disease mechanisms, processes such as RNA expression, splicing, translation, and stability regulation have become new targets for disease intervention. RNA has provided new therapeutic modalities for metabolic diseases, genetic disorders, and cancer patients, resulting in several innovative drugs.

MCE R&D team collected small molecules targeting RNA from the PDB, R-BIND, ROBIN, and internal database as the positive dataset, and non-targeting RNA small molecules from ROBIN as the negative dataset. Based on the GeminiMol pre-trained model, we encoded the molecules and calculated over 1700 molecular descriptors using Mordred as inputs for the model. Subsequently, we employed 13 deep learning models to learn from the data. All of which yielded good training results, with AUROCs greater than 0.75. Ultimately, we selected the Finetune model to screen HY-L901P, which exhibited the best classification performance, achieving an AUROC of 0.82 and a prediction accuracy of 0.76. We then applied filtering based on StaR rules (with at least two of the following properties: cLogP ≥ 1.5, Molar Refractivity ≥ 4, Relative Polar Surface Area ≤ 0.3) to obtain a library containing approximately 5,000 small molecule compounds targeting RNA. This library serves as a valuable tool for screening small molecules that interact with RNA.

Ion channels are key proteins on the cell membrane that regulate the flow of ions across membranes. They participate in nearly all physiological processes, including nerve conduction, muscle contraction, heart rhythm, and pain perception. Abnormalities in their function can lead to various serious diseases such as arrhythmia, epilepsy, hypertension, neuropathic pain, and cancer. Therefore, ion channels are highly valuable drug targets—over 15% of approved drugs target ion channels currently, demonstrating their irreplaceable therapeutic value in cardiovascular, neurological, and analgesic fields.

MCE has collected a library of over 5,000 reported ion channel-related bioactive compounds targeting major sites such as Na+ channels, K+ channels, Ca2+ channels, GABA receptors, iGluRs, and others. Using AI models, these compounds are characterized through both 2D representations (molecular fingerprints, pharmacophores) and 3D representations (3D conformation) to screen for a collection of lead-like compounds highly similar to known active molecules. Additionally, an hERG channel prediction algorithm integrating XGB and ISE mapping strategy is employed to assess and exclude potential cardiotoxicity in the library.. This step significantly reduces safety risks in subsequent screenings, particularly for ion channel drug development related to cardiovascular systems (e.g., Nav1.5, Cav1.2), effectively minimizing failures due to hERG inhibition and serving as a valuable tool for ion channel drug screening.

Drug-induced liver injury (DILI; also known as drug-induced hepatotoxicity) is caused by medications (prescription or OTC), herbal and dietary supplements (HDS), or other xenobiotics that result in abnormalities in liver tests or in hepatic dysfunction that cannot be explained by other causes. Drugs are an important cause of liver injury. Drug-induced hepatic injury is the most common reason cited for withdrawal of an approved drug.

DILI is thought to occur via several different mechanisms. Among these are direct impairment of the structural (e.g., mitochondrial dysfunction) and functional integrity of the liver; production of a metabolite that alters hepatocellular structure and function; production of a reactive drug metabolite that binds to hepatic proteins to produce new antigenic drug-protein adducts, which are targeted by hosts’ defenses (the hapten hypothesis); and initiation of a systemic hypersensitivity response (i.e., drug allergy) that damages the liver.

MCE Drug-induced Liver Injury (DILI) Compound Library contains a unique collection of 641 hepatotoxicity causing compounds and is a powerful tool to research DILI and other drug toxicities. This library can be used to understand the mechanisms of DILI, identify biomarkers for early DILI prediction, and allow timely recognition during drug development, thus finally achieving successful DILI prevention and assessment in the pre-marketing phase.