virtual

Molecular Glue Virtual Library is constructed using generative AI technology, integrating the structural features, activity data of known molecular glues, and interaction information of ternary complexes (target protein-E3-molecular glue). Endowed with structural novelty, drug-likeness, diversity and synthesizability, it is applicable to molecular glue-based AI drug screening and large-scale virtual screening.

MCE builds this library based on high-quality molecular building blocks by virtue of robust computing power, coupled with rigorous reaction rules and optimized compound generation strategies. To ensure library quality, molecules with high synthetic difficulty, poor drug-likeness, PAINS and other undesirable molecules are excluded first. Subsequently, scaffold-based compound analysis is performed to screen drug-like diverse molecules for synthesizability evaluation; those with excessively high synthetic difficulty are removed, ultimately forming a large-scale molecular glue virtual library with structural diversity, synthesizability and drug-likeness.

Compounds in the library can be synthesized in only 1-2 chemical reaction steps. With MCE’s experienced chemical synthesis team, custom synthesis of different scales from milligram to kilogram can be easily achieved to meet diverse customer needs.

MegaUni 50K Virtual Diversity Library consists of 50,000 novel, synthetically accessible, lead-like compounds. With MCE's 40,662 Building Blocks, covering around 273 reaction types, more than 40 million molecules were generated. Based on Morgan Fingerprint and Tanimoto Coefficient, molecular clustering analysis was carried out, and molecules closest to each clustering center were extracted to form a drug-like and synthesizable diversity library. The selected 50,000 drug-like molecules have 46,744 unique Bemis-Murcko Scaffolds (BMS), each containing only 1-3 compounds. This diverse library is highly recommended for virtual screening and novel lead discovery.

Macrocyclic compounds (≥12-atom cyclic small molecules/peptides) have unique physicochemical properties. They form preorganized conformations with high binding affinity/selectivity, target traditional small-molecule-inaccessible proteins, and bridge small-molecule drugs and biological agents. As key protein phosphorylation enzymes, kinases are linked to tumors, COPD, etc., and are critical therapeutic targets. Traditional small-molecule kinase inhibitors lack selectivity, causing off-target toxicity, low bioavailability, and acquired resistance. Macrocycles’ semi-rigid structure restricts conformations, boosts binding selectivity, optimizes pharmacokinetics, and makes macrocyclization a core kinase inhibitor optimization strategy.

Thousands of bioactive macrocycles were curated from ChEMBL. Via Transformer, macrocyclization was converted into a chemical language translation task, enabling end-to-end macrocycle generation from linear precursors with simplified inputs. Macformer achieves efficient, automated linear molecule macrocyclization via deep learning; generated macrocycles have diversity, novelty, biocompatibility, and cover broader chemical space.

MCE collected thousands of marketed/clinical kinase inhibitors, using their fragments for macrocyclization to generate derivatives. After evaluating synthetic accessibility and physicochemical properties, a million-scale virtual macrocyclic library was built for kinase-related virtual and AI-driven screening.

Macrocyclic compounds (≥12-atom cyclic small molecules/peptides) have unique physicochemical properties. They form preorganized conformations with high binding affinity/selectivity, target traditional small-molecule-inaccessible proteins, and bridge small-molecule drugs and biological agents. As key protein phosphorylation enzymes, kinases are linked to tumors, COPD, etc., and are critical therapeutic targets. Traditional small-molecule kinase inhibitors lack selectivity, causing off-target toxicity, low bioavailability, and acquired resistance. Macrocycles’ semi-rigid structure restricts conformations, boosts binding selectivity, optimizes pharmacokinetics, and makes macrocyclization a core kinase inhibitor optimization strategy.

Thousands of bioactive macrocycles were curated from ChEMBL. Via Transformer, macrocyclization was converted into a chemical language translation task, enabling end-to-end macrocycle generation from linear precursors with simplified inputs. Macformer achieves efficient, automated linear molecule macrocyclization via deep learning; generated macrocycles have diversity, novelty, biocompatibility, and cover broader chemical space.

MCE collected thousands of marketed/clinical kinase inhibitors, using their fragments for macrocyclization to generate derivatives. After evaluating synthetic accessibility and physicochemical properties, a million-scale virtual macrocyclic library was built for kinase-related virtual and AI-driven screening.

With MCE's 40,662 BBs, covering around 273 reaction types, more than 40 million molecules were generated. Compounds which comply with Ro5 criteria were selected. Inappropriate chemical structures, such as PAINS motifs and synthetically difficult accessible, were removed. Based on Morgan Fingerprint, molecular clustering analysis was carried out, and molecules close to each clustering center were extracted to form this drug-like and synthesizable diversity library. These selected molecules have 805,822 unique Bemis-Murcko Scaffolds (BMS) with diversified chemical space. This library is highly recommended for AI-based lead discovery, ultra-large virtual screening and novel lead discovery.

COVID-19 poses a serious threat to people's health, and it is urgent to develop drugs to treat COVID-19 quickly. The screening of anti-COVID-19 drugs by using the clinical and approved compounds can greatly shorten the research and development cycle. In addition, the virtual screening technology can effectively narrow the scope of screening and improve the screening efficiency in the pre-screening of new drugs.

Taking advantage of our virtual screening, we conduct virtual screening of approved compound library and clinical compound library based on the 3CL protease (PDB ID: 6LU7), Spike Glycoprotein (PDB ID: 6VSB), NSP15 (PDB ID: 6VWW), RDRP, PLPro and ACE2 (Angiotensin Converting Enzyme 2) structure. We design a unique collection of 1,380 compounds which may have anti-COVID-19 activity. Anti-COVID-19 Compound Library will be a powerful tool for screening new anti-COVID-19 activity drugs.

On May 15, 2024, "Dimerization and antidepressant recognition at noradrenaline transporter" was published online by Nature. The research findings were an effort from Shanghai Institute of Materia Medica, Chinese Academy of Sciences. This study unraveled the important neural system target - the noradrenaline transporter (NET), obtaining the binding modes of human NET homodimers with the natural substrate norepinephrine (NE) and six selective antidepressants. It laid an important theoretical foundation for understanding the physiological regulation mechanisms of NET and other monoamine transporters.

The Norepinephrine Transporter (NET) Compound Library is obtained by computer-aided virtual screening based on the HY-L901 compound library . The specific screening process includes molecular docking screening, key pharmacophore screening, and CNS-MPO screening, which can be used for new drug discovery targeting the noradrenaline transporter.

A diversity compound library contains 1,000,000 compounds with drug fragments. Each compound has at least one drug fragment. These selected molecules have 702,902 Bemis-Murcko Scaffolds (BMS) with drug-like chemical space. This library is highly recommended for AI-based lead discovery, ultra-large virtual screening and novel lead discovery.

PD-1/PD-L1 are key immune checkpoint targets that suppress T-cell-mediated anti-tumor immunity, representing a major focus in cancer immunotherapy. While antibody drugs dominate the clinic, they are limited by administration challenges and immune-related side effects. Small-molecule PD-1/PD-L1 inhibitors, with oral availability, good tissue penetration and low cost, have emerged as a promising next-generation strategy.

A PD-1/PD-L1 lead-like library was built via a five-step virtual screening process. After collecting 8,947 inhibitors from BindingDB and PubChem and filtering by activity and duplicates, AI similarity screening was performed using GeminiMol. Key pharmacophores were extracted from the PPI interface of co-crystal structures, and molecular was screened via a pharmacophore model, effectively enhancing target activity.

Containing 10,000 structurally diverse and drug-like molecules well-matched to the PD-L1 pocket, the library supports virtual docking, high-throughput screening and hit discovery, enabling efficient and rapid development of small-molecule immunotherapies.