We're excited to announce that the groundbreaking IBM MAMMAL (Molecular Aligned Multi-Modal Architecture and Language) is now available through our free public API at Cure Cancer With AI! Developed by IBM Research, MAMMAL is a multi-modal biomedical foundation model designed to accelerate research in cancer and drug discovery. This innovative model processes data across proteins, small molecules, and gene/omics information, making it a versatile tool for researchers and developers alike.

What is MAMMAL and Why Does it Matter?

MAMMAL stands for Molecular Aligned Multi-Modal Architecture and Language, and it represents a significant advancement in the field of biomedical AI. By integrating various biological modalities, MAMMAL provides a unified framework for understanding complex biological interactions. This is particularly important in cancer research and drug discovery, where understanding the interactions between proteins and small molecules can lead to more effective treatments.

The model utilizes a structured prompt syntax to perform classification, regression, and generation tasks, making it suitable for a wide range of applications. With the ability to analyze extensive datasets, MAMMAL helps researchers uncover vital insights that can lead to breakthroughs in understanding disease mechanisms and developing new therapies.

The Research Behind MAMMAL

MAMMAL's capabilities are backed by rigorous research, as detailed in the npj Drug Discovery paper published in Nature. Trained on approximately 2 billion biological samples, MAMMAL was evaluated on 11 diverse drug-discovery tasks. It achieved state-of-the-art results on 9 of these tasks, demonstrating its efficacy and reliability in predicting biological interactions.

This extensive training allows MAMMAL to analyze and predict interactions with remarkable accuracy, making it an invaluable resource for researchers in the field of drug discovery and cancer research.

MAMMAL in the Cure Cancer With AI API

We are thrilled to provide MAMMAL's powerful predictions through three API endpoints available on Cure Cancer With AI. All endpoints are POST requests and return results in JSON format. Below are the details of each endpoint, including the inputs and outputs.

• Protein–Protein Interaction (PPI)

  POST /api/v1/mammal/ppi

  This endpoint predicts whether two proteins interact. You need to provide two amino-acid sequences:

  • protein_a: The first protein's sequence.
  • protein_b: The second protein's sequence.

  The output will return a binding-affinity class label:

  • 1: Interacting
  • 0: Non-interacting

• Drug–Target Interaction (DTI)

  POST /api/v1/mammal/dti

  This endpoint predicts how strongly a drug binds to a target protein. You will need to provide:

  • target_seq: The amino-acid sequence of the target protein.
  • drug_seq: The drug in SMILES notation.

  The output will return a predicted pKd value (−log₁₀ Kd), where a higher value indicates stronger predicted binding affinity.

• ClinTox Clinical-Trial Toxicity

  POST /api/v1/mammal/clintox

  This endpoint predicts the potential toxicity of a compound in clinical trials. You need to provide:

  • smiles: The compound in SMILES notation.

  The output includes a toxicity prediction:

  • 1: Toxic or likely to fail trials
  • 0: Not toxic

  Additionally, a raw score will be provided for further analysis.

Start Using It for Free

The MAMMAL inference capability is now available for free on Cure Cancer With AI! Anyone can create a free API key at /api-keys. Our API allows up to 100 requests per hour per key, making it accessible for both individual researchers and larger projects. To read the full documentation, including parameters and code samples, visit /developers.

Conclusion

The addition of IBM MAMMAL AI to our public API represents a significant step forward in making advanced biomedical AI tools accessible to everyone. With MAMMAL now available for free on Cure Cancer With AI, researchers can leverage its powerful predictions to enhance their work in cancer and drug discovery.

As always, please remember that while these predictions can provide valuable research signals, they are not a substitute for medical advice, diagnosis, or treatment. The MAMMAL API serves as an experimental tool to aid in biomedical research.

To dive deeper into practical AI-for-cancer-research updates, explore our latest blog posts, learn more about our mission, and see how you can support ongoing work on our donations page.

As this field keeps evolving, platforms like curecancerwithai.com can help people follow progress, understand how AI is being applied in oncology, and stay connected to the broader mission.