At Cure Cancer With AI, we are thrilled to announce that we now offer access to IBM's groundbreaking MAMMAL AI through our free public API. MAMMAL, which stands for "Molecular Aligned Multi-Modal Architecture and Language," is an innovative biomedical foundation model developed by IBM Research. Designed specifically for cancer and drug discovery research, MAMMAL excels in processing and analyzing complex biological data across multiple modalities, including proteins, small molecules, and gene/omics data.

But what does it mean to be a multi-modal biomedical foundation model? Simply put, MAMMAL integrates various types of biological data into a unified framework, enabling researchers to gain deeper insights into molecular interactions and drug efficacy. This capability is vital for accelerating drug discovery and enhancing our understanding of cancer biology, ultimately paving the way for more effective therapies.

Research Behind MAMMAL

The capabilities of MAMMAL are backed by rigorous research, as detailed in the paper published in npj Drug Discovery (Nature). MAMMAL was trained on approximately 2 billion biological samples and evaluated across 11 diverse drug-discovery tasks. Impressively, it achieved new state-of-the-art results on 9 of these tasks, while performing comparably to the best existing models on the remaining 2. This extensive training and evaluation make MAMMAL a powerful tool for researchers looking to leverage AI in their drug discovery efforts.

MAMMAL in the Cure Cancer With AI API

We are excited to integrate MAMMAL's capabilities into our API, making powerful predictions accessible to everyone, for free! The API includes three endpoints that enable users to explore protein interactions, drug-target interactions, and clinical toxicity predictions.

• Protein–Protein Interaction (PPI): Using the endpoint /api/v1/mammal/ppi, you can input two amino-acid sequences (e.g., protein_a and protein_b). The API will return a binding-affinity class label of "1" if the proteins interact and "0" if they do not. This information is crucial for understanding cellular processes and signaling pathways.
• Drug–Target Interaction (DTI): Access the endpoint /api/v1/mammal/dti to predict how well a drug binds to a target protein. By providing the target protein's amino-acid sequence and the drug's SMILES notation, the API will return a predicted pKd value. A higher pKd indicates a stronger predicted binding, helping researchers identify promising drug candidates.
• ClinTox Clinical-Trial Toxicity: The endpoint /api/v1/mammal/clintox allows users to assess the potential toxicity of a compound. By entering a compound in SMILES notation, the API will predict whether it is likely to be toxic or not, returning a prediction of "1" (toxic/likely to fail trials) or "0" (not toxic), along with a raw score. This information is invaluable for making informed decisions in preclinical development.

All three endpoints are designed to be user-friendly, allowing researchers to harness the power of MAMMAL without needing extensive expertise in machine learning. Each call to the API is authenticated and takes approximately 60 seconds due to the CPU-bound nature of the model. Plus, you can make up to 100 requests per hour per API key—absolutely free!

Start Using It for Free

Getting started is easy! Simply create a free API key at /api-keys and dive into our comprehensive documentation at /developers. The docs provide all the details you need, including parameters, code samples, and best practices for using MAMMAL's endpoints.

Remember, the MAMMAL inference capabilities are freely available on Cure Cancer With AI, empowering researchers and developers to explore and innovate in the field of drug discovery.

Conclusion

In summary, we are proud to provide free access to the IBM MAMMAL AI through our public API at Cure Cancer With AI. This powerful tool can help researchers make significant strides in understanding protein interactions, drug-target relationships, and clinical toxicity predictions. Please note that while MAMMAL provides valuable research signals, it is not intended to replace medical professionals or provide clinical advice. Always consult with a qualified expert for medical 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.