MolPE 分子势能
中EN
Request a trial

Your AI colleague in molecule design

Not a tool, but your AI research colleague: it breaks down tasks, plans experiments, calls tools and optimizes with you. For protein, antibody and peptide research, MolEX brings the research assistant, molecular models and scientific knowledge into one workspace.

Chain H · heavy chain

Chain K · heavy chain

Chain L · light chain

Chain M · light chain

MolEX AgentPDB 1HZH

PDB 1HZH is loaded. What would you like to know about this structure?

Start with one of these questions:

4 protein chains: H and K are IgG1 heavy chains, L and M are light chains. Each heavy chain pairs with a light chain to form a Fab arm, and the heavy chains’ constant regions join below as the Fc: an intact human IgG1 antibody, b12.

From the N-terminus, each heavy chain runs VH, CH1, hinge, CH2, CH3, and each light chain VL, CL. At the tip of each Fab arm, 6 CDR loops form the antigen-binding site, now marked on the structure. CDR H3 is the longest, at 18 residues.

1HZH is itself the binder, not a target: it is the antibody b12. Its target is the HIV-1 envelope glycoprotein gp120, where b12 binds the CD4-binding site (the complex is PDB 2NY7). To assess a target, load the gp120 structure and ask again.

Watch MolEX run a real analysis

It starts from one research goal: MolEX breaks the task down, calls tools, asks the scientist to confirm parameters, schedules compute, and delivers structures, metrics and a report.

Read the changelog

MolEX’s three core capabilities

  • Multi-agent collaboration

    A hierarchical decision mechanism models how a researcher thinks a problem through, so complex tasks run to completion on their own.

    • Task decomposition and planning
    • Agents executing together
    • Hierarchical decisions with reflection

  • Knowledge and tools

    Expert knowledge meets upstream and downstream toolsets; tools evolve on their own, and models grow tuned to each team.

    • In-house and open-source models
    • Targeted learning from private knowledge
    • Tool chains as plug-ins

  • Memory and reasoning

    Self-reflective iterative optimization and global–local memory control make retrieval faster and tasks more stable.

    • Global–local layered memory
    • Self-reflective iteration
    • A traceable research record

Research collaboration starts with a question

See how MolEX connects research activities, from understanding sequences to exploring molecular structures.

Understand intent, execute autonomously

  • Intelligent task breakdown and planning
  • Multi-agent collaborative execution
  • Hierarchical decisions with reflective refinement

3D structure editing, in sync with the agent

  • Residue selection and candidate comparison
  • Design Copilot interprets in sync
  • Collaborative annotation by members

Empowered by knowledge and tools

  • Targeted learning from a private knowledge base
  • In-house and open-source models integrated
  • Toolchain extended through plug-ins

Build your own skills and workflows

  • Drag-and-drop workflow nodes
  • Upload and manage Skills
  • Team SOPs as executable chains

Readable output, traceable process

  • Multiple sequence alignment view
  • One-click export of data, figures and reports
  • Layered memory, a traceable process
  • AlphaFold 3structure
  • ESM-2language
  • RFdiffusiondesign
  • FoldXevaluation
  • Boltz-2structure
  • ProteinMPNNdesign
  • Rosettaevaluation
  • AntiBERTylanguage
  • Chai-1structure
  • BindCraftdesign
  • ThermoMPNNevaluation
  • ESM Clanguage
  • ColabFoldstructure
  • PepMLMdesign
  • FreeSASAevaluation
  • ProtGPT2language
  • IgFoldstructure
  • DiffDockstructure
  • GearBindevaluation
  • AbLang-2language
  • ABodyBuilder2structure
  • OpenFoldstructure
  • OpenStructureevaluation
  • ESMFoldstructure
  • HADDOCKstructure
  • ImmuneBuilderstructure
  • AlphaFold 2structure

All models in the model hub

Fab–TNFα binding review
Continuous playback

Agent session: task breakdown, sub-agents in parallel, a plan revised after reflection

Eight supporting functions for the whole research workflow

From file management to custom workflows: one place for every research scenario.

  • Workspace

    Files and tasks across the platform in one place: sequences, structures, literature and agent outputs, clearly organized.

  • Knowledge base

    Upload your own data into a dedicated database; agents learn from it for sharper accuracy in your field.

  • Smart search

    Search targets, pipelines, literature and other research material, linked to your private knowledge base.

  • Model hub

    The platform’s in-house and open-source computational models, integrated in one place.

  • Scheduled tasks

    Batch computations and data updates run in the background, unattended.

  • Plugin market

    Research plugins for data analysis, paper writing and more, installed as needed.

  • Workflow

    Pro users define and upload skills, turn company SOPs into agent-executable chains, and make agents stronger on specific tasks.

  • Omics atlas

    Across cells, genes and proteins, explore the structures and connections in multi-omics and single-cell data.

From amino acid sequence to molecular function

Connect computational analysis and design exploration to help teams focus on directions for further validation.

  • Epitope research

    Explore molecular recognition and binding interfaces using structural information.

    Structural understanding

  • CDR design

    Design and evaluate antibody variable regions around functional objectives.

    Sequence modelingGenerative exploration

  • Affinity optimization

    Connect sequence changes with binding behavior to compare and refine candidate designs.

    Sequence modelingConnected research

  • Developability research

    Bring stability, solubility and related needs into the design stage.

    Multi-property assessment

  • De novo design

    Explore protein and peptide sequence space through generative methods.

    Generative explorationMulti-property assessment

  • Structural research

    Use computational models to study how sequence relates to conformation.

    Structural understandingSequence modeling

Computation informs research decisions. Molecular performance requires experimental validation.

Working together on research

Bring your research objectives to MolEX

Book a consultation to explore how MolEX can fit your research.

Contact us