We are excited to announce the availability of the IBM MAMMAL (Molecular Aligned Multi-Modal Architecture and Language) biomedical foundation model through our free public API at Cure Cancer With AI. Developed by IBM Research, MAMMAL represents a significant advancement in the field of drug discovery AI, leveraging a multi-modal approach that encompasses proteins, small molecules, and gene/omics data. This model is designed to facilitate research in cancer and drug discovery by providing valuable predictions that can aid in understanding complex biological interactions.

What is MAMMAL and Why is it Important?

MAMMAL is an innovative multi-task model that integrates various biological data types, including proteins, small molecules, and single-cell gene/omics information, into a unified sequence framework. This means that MAMMAL can analyze and interpret data from different sources simultaneously, which is crucial for accurate drug discovery and understanding the biological mechanisms of diseases like cancer.

The ability to work across multiple modalities allows researchers to gain insights that are not possible with traditional, single-modality models. This is particularly important in the context of cancer research, where the interactions between different biological components are complex and multifaceted. MAMMAL’s architecture enables it to provide predictions that can inform drug development, identify potential targets for therapies, and ultimately contribute to better treatment strategies.

Research Behind MAMMAL

The capabilities of MAMMAL are grounded in extensive research, as detailed in the paper published in npj Drug Discovery (Nature). This research involved training the model on approximately 2 billion biological samples, making it one of the most comprehensive biomedical models to date. MAMMAL was evaluated on 11 diverse drug-discovery tasks, achieving state-of-the-art results on 9 of them, and comparable performance on the remaining 2. This rigorous evaluation highlights MAMMAL's effectiveness and reliability as a tool for researchers in the biomedical field. You can read more about the methodology and findings in the full paper [here](https://www.nature.com/articles/s44386-026-00047-4).

MAMMAL in the Cure Cancer With AI API

We are thrilled to offer three powerful MAMMAL API endpoints through our platform, allowing users to access its predictions for free. Each endpoint is designed for specific tasks, providing researchers with valuable insights into protein interactions, drug-target interactions, and clinical toxicity. Here’s a closer look at each endpoint:

• Protein–Protein Interaction (PPI):

  Endpoint: /api/v1/mammal/ppi

  This endpoint takes two amino-acid sequences (designated as protein_a and protein_b) as input. It returns a binding-affinity class label of "1" (indicating that the proteins interact) or "0" (indicating no interaction). This prediction can help researchers understand how proteins communicate and function together within biological systems.

• Drug–Target Interaction (DTI):

  Endpoint: /api/v1/mammal/dti

  For this endpoint, users provide a target protein amino-acid sequence (denoted as target_seq) and a drug in SMILES notation (denoted as drug_seq). The output is a predicted pKd value (−log₁₀ Kd), where a higher value suggests a stronger predicted binding affinity between the drug and the target protein. This information is crucial for drug discovery efforts, helping to identify promising candidates for further study.

• ClinTox Clinical-Trial Toxicity:

  Endpoint: /api/v1/mammal/clintox

  This endpoint requires a compound to be inputted in SMILES notation (denoted as smiles). It returns a toxicity prediction, with a result of "1" indicating that the compound is toxic or likely to fail clinical trials, and "0" indicating that it is not toxic. Additionally, a raw score is provided to give more context to the prediction. This endpoint is invaluable for researchers evaluating the safety of potential new drugs.

Start Using MAMMAL for Free

We invite researchers, developers, and enthusiasts to explore the capabilities of MAMMAL through our public API, available free of charge on Cure Cancer With AI. To get started, simply create a free API key at /api-keys. With this key, you can make up to 100 requests per hour, allowing you to integrate MAMMAL's powerful predictions into your research and applications. Full documentation, including parameters and code samples, is available at /developers.

Conclusion

The launch of MAMMAL predictions through our free public API marks a significant step forward in leveraging AI for biomedical research. Researchers now have access to advanced prediction capabilities that can enhance their understanding of complex biological interactions and contribute to the development of new therapies. We encourage you to explore these endpoints and take advantage of this free resource at Cure Cancer With AI.

As a reminder, while MAMMAL provides valuable predictive insights, these predictions are intended as research signals and should not be interpreted as medical or clinical advice. Always consult medical professionals for clinical decisions.

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.

Cure Cancer With AI is an educational research and information platform. It does not provide medical advice, diagnosis, or treatment recommendations; always discuss care decisions with a qualified healthcare professional.