Molecular Glue Database: A New Frontier in Drug Discovery

Introduction

In drug discovery, breakthroughs rarely come in neat packages. They often emerge from unexpected mechanisms that challenge how we think about biology. The molecular glue database is one such development. At first glance, it may seem like just another online resource. But for researchers in targeted protein degradation—a field reshaping modern pharmacology—MolGlueDB is a quiet revolution.

This article explores what makes MolGlueDB stand out, how it works, and why it matters. By the end, you’ll understand why molecular glues are more than a scientific curiosity, and how this database could accelerate real-world therapeutic advances.

What Are Molecular Glues?

Molecular glues (MGs) are small molecules that promote interactions between a target protein and an E3 ubiquitin ligase. Unlike bifunctional PROTACs, which are designed with two distinct binding arms, molecular glues operate with a simpler structure. They stabilize a protein–protein interaction that otherwise would not occur, leading the cell’s degradation machinery to break down the target protein.

This approach is attractive because:

• MGs are smaller, more drug-like molecules.
• They can reach previously “undruggable” proteins.
• They offer a cleaner mechanism of action compared to traditional inhibitors.

For example, lenalidomide—a well-known immunomodulatory drug—works as a molecular glue by recruiting cereblon (an E3 ligase) to degrade transcription factors like Ikaros and Aiolos. This success story triggered a wave of interest in identifying new molecular glues for oncology, neurology, and beyond【academic.oup.com】.

Introducing MolGlueDB: Molecular Glue Database

Launched in 2025, MolGlueDB (accessible at molgluedb.com) is the first comprehensive database dedicated entirely to molecular glue degraders. It consolidates knowledge from 241 publications between 2001 and 2025, covering:

• 1,629 unique molecular glues
• 94 target proteins
• 28 recruiting proteins
• 19 clinical-stage compounds
• 1,840 total entries【academic.oup.com】

What sets MolGlueDB apart is not only the breadth of its data but also its usability. Researchers can search by molecular ID, chemical structure, target, recruiting protein, or even draw a structure directly.

How to Search the Molecular Glue Database

The search interface is intuitive yet powerful. You can choose from multiple entry points:

• MGDB ID: if you already know the database identifier.
• SMILES string: for chemical structure searches.
• Target: to identify glues against specific proteins.
• Recruiting Protein: to focus on ligases like cereblon or DDB1.
• Name: for compounds with common identifiers.
• Draw Structure: for chemists who want to sketch a scaffold.

Once a query is submitted, results appear in a tabular view, listing:

• Structure and MGDB ID
• Name and activity status
• Recruiting protein
• Primary and secondary targets
• Mode of action
• Research stage
• Pharmacophore and core
• Affinity values
• Degron type
• Structural data (PDB, CryoEM)
• Key physicochemical properties (molecular weight, tPSA, clogP, cLogS, H-bond donors/acceptors, rotatable bonds)

Each record includes a Details option, expanding into a full profile:

• Summary
• Molecular representations
• Calculated properties
• Target information
• Experimental physicochemical data
• Activity results
• Linked publications

This layered structure ensures both high-level browsing and deep exploration.

Filtering and Advanced Options

MolGlueDB includes filters that help refine large result sets. Available filters include:

• Research stage (discovery, preclinical, clinical)
• Target and recruiting protein
• IsActive (activity evidence)
• Mode of action
• Pharmacophore and core scaffold
• Degron type
• Complex structure availability
• Molecular weight distribution
• tPSA, clogP, and other physicochemical ranges

For instance, a researcher could filter for “clinical-stage” compounds recruiting cereblon, with a molecular weight under 500 Da and a tPSA under 120 Ų—quickly narrowing candidates for oral bioavailability.

Practical Example: Using the Molecular Glue Database

Imagine a biotech scientist interested in BRD4 degradation, a hot target in oncology. By entering BRD4 in the Target field, MolGlueDB returns all glues linked to BRD4 degradation.

The scientist can then filter by:

• Research stage: to focus on molecules with preclinical or clinical evidence.
• Recruiting protein: e.g., cereblon-based degraders.
• Mode of action: direct binders vs. allosteric modulators.

From there, they could examine detailed records including ternary EC50 values, structural PDB entries, and linked activity data. This workflow illustrates how MolGlueDB bridges raw chemical data with biological context.

Why This Database Matters

MolGlueDB is more than a catalog. It provides:

1. Centralization of knowledge – integrating scattered literature into a single searchable platform.
2. Support for rational design – allowing chemists and biologists to cross-compare scaffolds, targets, and properties.
3. Clinical relevance – tracking molecules in the pipeline, not just theoretical candidates.
4. Structural insights – linking chemical entries to CryoEM and PDB structures.
5. Community value – as an open-access resource, it democratizes drug discovery tools.

Given the rising interest in targeted protein degradation, MolGlueDB could accelerate not only academic research but also industrial discovery pipelines.

Challenges and Limitations

While MolGlueDB is impressive, it has natural limits:

• The field is young; data coverage, especially for non-cereblon ligases, remains sparse.
• Not all compounds have robust activity data or structural validation.
• Integration with other drug discovery databases (e.g., ChEMBL, BindingDB) is currently limited.

These caveats do not diminish its utility but highlight where future updates could enhance cross-platform connectivity.

Conclusion

The molecular glue database is not a flashy tool—it’s a quietly radical one. By pulling together over two decades of scattered data, MolGlueDB enables researchers to move faster, ask better questions, and potentially design the next generation of therapies against diseases once thought “undruggable.”

For anyone working in drug discovery, structural biology, or chemical informatics, MolGlueDB is a resource worth bookmarking. It also stands as a case study in how science databases can transform niche fields into global movements.

Explore more resources like this in our dedicated section on Science Databases, including the MetaGRoot Database on Plant Root Microbiomes and the Localizatome Database.

Sources

• Chen et al., MolGlueDB: an online database of molecular glues, Nucleic Acids Research (2025). Oxford Academic
• MolGlueDB official site