molecular design

Targeted Protein Degradation (TPD) is a novel and promising approach to drug development. It shows great potential for targeting proteins traditionally considered "undruggable" due to the lack of enzymatic function and absence of binding sites by tagging them for degradation or recruiting natural degradation mechanisms.

Molecular glues are a type of small-molecule degraders that primarily induce novel interactions between E3 ubiquitin ligases and target proteins, forming ternary complexes that lead to protein ubiquitination and subsequent proteasomal degradation. Compared with PROTACs, molecular glues generally have lower molecular weights, higher cell permeability, and better drug-like properties. Additionally, the design of molecular glues is relatively simple, without the requirements for complex linkers and ligand optimization. As a result, molecular glues have gradually emerged as a promising therapeutic approach for various diseases.

Multiple types of molecular glues have been reported previously. Analysis of co-crystal complex structures reveals that CRBN-related molecular glues are more versatile. Therefore, MCE researchers select active molecules related to these targets as probes for artificial intelligence (AI) screening.Subsequently, molecular docking technology was used to verify whether the screened molecules retained the key pharmacophore features. Ultimately, we obtained 317 molecular glue analogs, and these compounds serve as powerful tools for the research of molecular glues.

CRBN, namely cereblon, is the substrate recognition subunit of the E3 ubiquitin ligase complex in the ubiquitin-proteasome system. A CRBN ligand library refers to a collection of numerous fragments that can specifically bind to the CRBN protein.

These ligands are mostly designed based on validated CRBN-binding warheads and modified through AI-driven molecular generation optimization systems. They not only include classic lenalidomide-derived structures but also cover novel non-lenalidomide scaffolds. After drug-likeness filtering, these ligands exhibit structural diversity and favorable druggable properties. They can be further optimized and modified to facilitate the development of novel molecular glue degraders, accelerate the discovery of molecular glues that induce interactions between CRBN and new substrate proteins, and enable the exploration of novel CRBN substrates for identifying previously unknown CRBN-binding proteins.

MCE compiles 125 fragments that can specifically bind to the CRBN protein, with molecular weights ranging from 200 to 500. Compounds developed based on the library ligands target multiple disease targets such as cancer and autoimmune diseases, further advancing the development of Molecular Glues and PROTACs therapeutic agents.

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.

Fragment-based drug discovery (FBDD) is well suited for discovering both drug leads and chemical probes of protein function. 3-dimensionality (3D) diversity is pivotal because the molecular shape is one of the most important factors in molecular recognition by a biomolecule. There is a developing appreciation that 3D fragments could offer opportunities that are not provided by 2D fragments.

MCE 3D Diverse Fragment Library consists of 5,400 non-flat fragment-like molecules (average Fsp3 value 0.58). More than 4,700 fragment compounds contain at least one chiral center in the structure. The key concepts that underlie the library design were 3D shape, structural diversity, reactive functionality and fragment-like. This 3D Diverse Fragment Library brings higher fragment hit optimization and increases the likelihood to find innovative hits in FBDD.

Unnatural amino acids (UAAs), also referred to as non-canonical amino acids (ncAAs) or non-proteinogenic amino acids, are a class of amino acids that are distinct from the 20 standard natural amino acids. They can be obtained through chemical synthesis, biosynthesis, and other approaches, with structural diversity far exceeding that of natural amino acids. UAAs are mainly including naturally occurring non-canonical amino acids, chemically synthesized amino acids, and biosynthetic amino acids, which provide a molecular basis for protein function design.

UAAs exhibit significant value in multiple fields. They can optimize the pharmacokinetic properties of peptide drugs and peptidomimetics, modify enzyme functions and endow them with new biological activities, thereby overcoming the limitations of traditional peptide drugs and expanding the chemical space . Meanwhile, UAAs can serve as molecular probes to analyze protein-protein interactions and investigate the regulatory mechanisms of protein functions.

MCE has compiled a UAAs Fragment Library comprising nearly a thousand unnatural amino acid fragments with extensive coverage of chemical space and enhanced structural diversity. This compound library can be widely applied in peptide synthesis, drug design, and protein engineering.

Peptidomimetics are compounds whose essential elements (pharmacophore) mimic a natural peptide or protein in 3D space and which retain the ability to interact with the biological target and produce the same biological effect. Peptidomimetics are designed to circumvent some of the problems associated with a natural peptide: e.g. stability against proteolysis (duration of activity) and poor bioavailability. Certain other properties, such as receptor selectivity or potency, often can be substantially improved. The design and synthesis of peptidomimetics are most important because of the dominant position peptide and protein-protein interactions play in molecular recognition and signaling, especially in living systems. Hence mimics have great potential in drug discovery.

MCE Peptidomimetic Library contains 370 compounds including peptoid, α-helix mimetics, β-turn/sheets mimetics, etc. This library is an indispensable tool of structure-activity relationships in drug discovery.

In drug discovery and development (R&D) area, target binding and druggability optimization are core processes. Among these attributes, high solubility is critical for a compound to achieve druggability, as it directly impacts the progress of drug R&D. Superior solubility ensures the rapid dissolution and uniform distribution of drug molecules in vivo, thereby enhancing bioavailability and effectively mitigating issues such as suboptimal efficacy, increased dosage requirements, or exacerbated toxic and side effects arising from insufficient solubility.

From the perspective of medicinal chemistry, high-solubility drug fragments serve as high-quality "molecular building blocks". Based on these fragments, lead compounds with potential druggability can be rapidly screened out, which significantly shortens the drug R&D cycle and reduces R&D costs. Meanwhile, the high-solubility drug fragment library can provide diverse options for drug development in different therapeutic areas, offer solutions for the solubility defects of existing clinical drugs, and facilitate the development of novel, highly effective targeted drugs with higher bioavailability and better safety profiles.

MCE has collected and compiled 2,527 experimentally validated small-molecule fragments with high solubility. These fragments can be directly used for drug molecular design, providing high-quality pre-validated solubility fragments that significantly improve the efficiency of lead compound screening and accelerate the progress of drug R&D.

ASINEX has elaborated a library of diverse macrocycles using an effective tool box of synthetic methods. The resulting scaffolds are novel, tremendously diverse, medchem-relevant, macrocyclic frameworks.

Macrocyles tend to be larger than traditional screening molecules which make them perfect discovery tools for targets with shallow or extended binding sites. At the same time, their unique character based on restricted flexibility and ability to form intra-molecular hydrogen bonds allows for design approaches effectively optimizing properties such asaqueous solubility and membrane permeability. Many of these macrocycles have been tested for aqueous and DMSO solubility with cut-offs applied at 10 mM in DMSO and 50 µM in PBS (pH 7.4) followed by PAMPA permeability assay.

“BioDesign” approach incorporates key structural features of known pharmacologically relevant natural products (e.g. alkaloids and other secondary metabolites) into synthetically feasible medicinal chemistry scaffolds. In order to identify the privileged pharmacophores, ring systems and linkers, we have carried out statistical analysis of structural features of natural products, marketed drugs, and drug candidates.

Saturated, fused ring, spiro, and bridged systems with a tendency towards multiple chiral centers are highly privileged among natural products and marketed drugs yet these structures are very poorly represented in commercial libraries. This library addressed this market need by incorporating these privileged elements into the design of novel synthetic molecules with high molecular framework diversity, multiple stereogenic centers (≥2), and degree of saturation (Fsp3 > 0.5).

Fragment-based drug discovery (FBDD) offers a strategic advantage by categorizing fragment hits according to their functional groups. This approach facilitates both the further optimization of these hits and the rational design of larger compounds through fragment combination. The amine functional group plays a vital role in drug development, as evidenced by its presence in many marketed drugs like Galantamine, Tacrine, and Rivastigmine. It is instrumental in enhancing solubility, improving bioavailability, and ensuring shelf-life stability—all critical factors for drug efficacy.

MCE offers a collection of 19,532 amine fragments for drug discovery. All of these compounds adhere to the Rule of Three (RO3) criteria for drug-likeness, which MCE offers a collection of 19,532 amine fragments for drug discovery, all of which stipulates a molecular weight ≤ 300 Da, ≤ 3 hydrogen bond donors, ≤ 3 hydrogen bond acceptors, and a cLogP ≤ 3.

Cysteine proteases (CPs), a key enzyme family regulating physiological metabolism and mediating pathological processes (such as abnormal bone resorption, tumour invasion, and pathogen infection), represent a core therapeutic target for developing specific inhibitors in disease intervention. Currently reported CP inhibitors primarily achieve their inhibitory function by precisely binding to CP active pockets (e.g., S1-S4 non-primed regions or S1'-S2' primed regions) and forming covalent/non-covalent interactions with the active site cysteine residues, providing clear structural references for the development of novel inhibitors.

This compound library, designed based on the core strategy of "similarity-based known active structures", contains over 200 cysteine protease inhibitors. Leveraging AI-driven molecular screening technology, it retains the critical pharmacological and shape features of reported CP inhibitors, serving as a specialized tool for efficiently discovering novel cysteine protease inhibitors.

Cosmetics are complex mixtures formulated through the rational blending and processing of various natural, synthetic, or extracted raw materials. The ingredients used in cosmetics are diverse in type and functionality. Based on their properties and applications, cosmetic ingredients can generally be classified into two main categories: base materials and auxiliary ingredients. The former constitutes the primary component of cosmetics, accounting for a significant proportion in formulations and serving as the key functional substances. Auxiliary ingredients, on the other hand, play roles in shaping the product, ensuring stability, or imparting color, fragrance, and other specific characteristics.Establishing a systematic ingredient database is of great significance for cosmetic research and development. By analyzing the physicochemical properties (such as oil-water partition coefficients and molecular weight distribution), biological activity mechanisms (such as antioxidant pathways and cellular signaling regulation), and compatibility of ingredients, the database can provide precise guidance for formulation design. This approach helps shorten the development cycle of cosmetic/skincare products and reduces trial-and-error costs.

MCE offers 622 types of cosmetic ingredient compounds, including antioxidants, humectants, emulsifiers, film-forming agents, and more.

DEL technology enables the simultaneous screening of millions or billions of compounds in a single tube by covalently linking each small molecule with a unique DNA sequence. Traditional DEL screening primarily focuses on identifying non-covalent binding molecules, where interactions with the target are reversible. In contrast, DNA‑encoded covalent library is an ultra‑high‑throughput screening library developed on the basis of conventional DNA‑encoded library technology. It incorporates controllable electrophilic covalent warheads capable of forming irreversible covalent bonds with amino acid residues at the active sites of target proteins, including Cys, Lys, Ser, Tyr, and others. This covalent binding enhances binding affinity, prolongs residence time at the target site, and has the potential to overcome challenges associated with traditional non-covalent inhibitors, such as drug resistance or off-target effects.

Each compound in the library contains both a binding domain and an electrophilic warhead. It first recognizes and binds to the target through non covalent interactions, and then forms a stable covalent bond with key amino acid residues to achieve irreversible inhibition. This library is specifically designed for the discovery of potent, long lasting, and highly selective covalent inhibitors, particularly for undruggable targets such as kinases, GPCRs, proteases, and mutant oncoproteins. Each molecule is uniquely labeled with a DNA barcode for molecular identification and sequencing decoding.

This library is an advanced and highly diverse collection, consists of 35 independent sub-libraries with a total scaleof 14 million compounds, It incorporates over 14 experimentally validated covalent warheads capable of targeting cysteine, lysine, arginine, aspartic acid and glutamic acid. This library is constructed with diverse drug like core scaffolds and integrated controllable covalent warheads, it features structural diversity, reaction spec

In contrast to the high conservation of conventional orthosteric sites, allosteric sites possess structural characteristics of low conservation, high hydrophobicity, weak polarity, confined spatial geometry, and dynamic cryptic properties. There is a significant difference between their core structures and orthosteric pockets — allosteric pockets are mostly dynamic grooves formed by protein conformational changes, subunit interface clefts, or shallow depressions, rather than the rigid "keyhole" structure of orthosteric sites. With looser spatial constraints, allosteric sites have the advantages of high selectivity and low off-target risk, and have become an important direction in new drug discovery.

Based on the dynamic, hydrophobic, and narrow-long spatial characteristics of allosteric pockets, MCE has performed targeted modification and screening of fragments. The screening criteria strictly conform to the requirements of allosteric binding: molecular weight is controlled at 120–280 Da (to meet the core needs of small molecules in fragment libraries and high derivatization), hydrogen bond donors (HBD ≤ 2), hydrogen bond acceptors (HBA ≤ 3), polar surface area (PSA = 30–80 Ų), rotatable bonds (≤ 2), moderate hydrophobicity (cLogP = 1–3.5), no strongly ionizable groups, and both appropriate rigidity and conformational flexibility to adapt to the dynamic changes of the pocket. Meanwhile, combined with the results of principal moment of inertia (PMI) analysis, fragments with high 3D diversity were obtained. Such fragments have good shape complementarity with allosteric pockets, ensuring that the fragments can smoothly enter the allosteric pockets and form stable binding, while providing room for subsequent optimization and derivation.

This library contains 1,800 structurally diverse fragment molecules with excellent drug-like properties, suitable for allosteric drug development and the design and optimization of allosteric sites. It combines the

Owing to the high conservation of orthosteric sites, conventional orthosteric drugs frequently suffer from poor subtype selectivity, off-target toxicity, and drug resistance, severely restricting their clinical application. In contrast, allosteric sites feature low conservation, high hydrophobicity, weak polarity, confined spatial geometry, and dynamic cryptic properties. These characteristics endow allosteric modulators with distinct advantages including high selectivity, functional tunability, and improved safety, making allosteric therapy a key direction in modern drug discovery.

MCE has curated nearly 1,000 structurally disclosed clinical-stage allosteric modulators. By analyzing allosteric protein–ligand complex structures from the PDB database, we extracted core pharmacophores and privileged scaffolds. Adopting a rational design strategy of “scaffold derivation + allosteric physicochemical filtering”, we performed secondary screening on the derived compounds strictly following the optimal physicochemical principles for allosteric binding based on universal allosteric pocket properties: molecular weight 300–500 Da, HBD ≤ 3, HBA = 3–8, PSA = 70–120 Ų, rotatable bonds ≤ 6, highly rigid scaffolds, cLogP = 1.0–3.8, and no strongly ionizable groups. The selected compounds exhibit high rigidity and shape complementarity, making them well-suited for targeting shallow, dynamic, and hydrophobic-dominated allosteric pockets.

This allosteric modulator library contains 4,315 structurally diverse, lead-like compounds dedicated to allosteric drug development, allosteric site targeting, and allosteric modulator screening. It is suitable for kinases, GPCRs, and other important drug targets. All compounds are analogs of clinical-stage allosteric modulators with a similarity score > 0.6, combining excellent druggability and allosteric binding potential. It provides a highly efficient tool for early-stage allosteric drug discovery.