We are excited to announce that the IBM MAMMAL (Molecular Aligned Multi-Modal Architecture and Language) model is now available through our free public API on curecancerwithai.com. Developed by IBM Research, MAMMAL is a cutting-edge multi-modal biomedical foundation model designed to enhance drug discovery and cancer research. By integrating data from proteins, small molecules, and gene/omics sources, MAMMAL takes a comprehensive approach to understanding the complex interactions within biological systems.

What is MAMMAL and Why Does it Matter?

MAMMAL stands out as a multi-modal model, meaning it processes and analyzes various types of biological data together, rather than in isolation. This unique capability is essential for cancer and drug-discovery research, where understanding the interplay between different biological elements can lead to more effective treatments and therapies. MAMMAL’s unified framework allows researchers to explore intricate relationships between proteins and small molecules, providing insights that can accelerate the drug discovery process.

The Research Behind MAMMAL

The foundation of MAMMAL's capabilities is detailed in the npj Drug Discovery paper published in Nature. This research highlights MAMMAL's impressive training on approximately 2 billion biological samples across multiple modalities, which enables it to excel in diverse tasks. Evaluated on 11 drug-discovery tasks, MAMMAL achieved state-of-the-art results in 9 of those tasks, demonstrating its potential to significantly impact the field of biomedical research. The model's proficiency in handling various data types allows for more nuanced predictions and analyses, ultimately driving forward our understanding of drug interactions and biological processes.

MAMMAL in the Cure Cancer With AI API

We are thrilled to integrate MAMMAL's powerful predictions into our API, making them accessible for free. The API includes three distinct endpoints, each designed to provide valuable insights to researchers and developers:

• Protein–Protein Interaction (PPI): This endpoint is accessed via POST /api/v1/mammal/ppi. By inputting two amino-acid sequences (protein_a and protein_b), users can receive a prediction indicating whether the two proteins interact. The output will return a binding-affinity class label of "1" for interacting proteins or "0" for non-interacting ones.
• Drug–Target Interaction (DTI): This endpoint is available at POST /api/v1/mammal/dti. Users need to provide the amino-acid sequence of a target protein (target_seq) and the drug's SMILES notation (drug_seq). In return, the API predicts the binding affinity of the drug to the target protein in the form of a pKd value (−log₁₀ Kd), where higher values indicate stronger predicted binding.
• ClinTox Clinical-Trial Toxicity: Access this endpoint through POST /api/v1/mammal/clintox. By submitting a compound in SMILES notation (smiles), users can obtain a toxicity prediction. The output will include a classification of "1" for compounds likely to be toxic or fail trials, and "0" for those predicted to be non-toxic, along with a raw score for further analysis.

These MAMMAL endpoints complement our existing data API, which includes access to research papers, news, blog posts, FDA approvals, clinical trials, and cross-dataset searches. All predictions generated using MAMMAL are research signals that can inform further studies, but they should not be construed as clinical advice.

Start Using It for Free

To leverage the power of MAMMAL for your research, simply create a free API key at /api-keys. With this key, you can make up to 100 requests per hour—perfect for exploring the capabilities of this advanced model. For detailed information about the parameters and usage, visit our documentation at /developers.

Conclusion

The integration of IBM MAMMAL into the Cure Cancer With AI API makes powerful predictions available for free, enabling researchers and developers to harness the potential of advanced biomedical AI. Dive into this innovative technology today and explore its capabilities by visiting curecancerwithai.com. Remember, while MAMMAL provides valuable research insights, these predictions are not medical or clinical advice and should be treated as experimental signals in the context of ongoing 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.

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.