At Cure Cancer With AI, we are excited to announce that we are now providing free access to the IBM MAMMAL (Molecular Aligned Multi-Modal Architecture and Language) biomedical foundation model through our public API. Developed by IBM Research, MAMMAL is a powerful multi-modal model that enables researchers to explore complex interactions within biological systems, making it a significant tool in the fields of cancer and drug discovery research.

What is MAMMAL?

MAMMAL stands out as a multi-modal biomedical foundation model that integrates various types of biological data, including proteins, small molecules, and single-cell gene/omics data. By representing these diverse modalities in a unified sequence framework, MAMMAL can perform a variety of tasks related to drug discovery, helping researchers uncover new insights into cancer treatment and therapeutic development.

The importance of a multi-modal approach in biomedical research cannot be overstated. By synthesizing data from different sources, MAMMAL can identify relationships and interactions that single-modal models might miss. This capability is crucial for understanding the complex biology of cancer and for advancing drug discovery efforts, as it allows for a more holistic view of the underlying mechanisms at play.

The Research Behind MAMMAL

The capabilities of MAMMAL are detailed in the research paper published in npj Drug Discovery (Nature), titled "MAMMAL — Molecular Aligned Multi-Modal Architecture and Language for biomedical discovery". This comprehensive study highlights that MAMMAL was trained on an impressive dataset of approximately 2 billion biological samples. The model has been evaluated across 11 diverse drug-discovery tasks, achieving state-of-the-art results in 9 of them and performing comparably on the remaining 2. This level of performance underscores MAMMAL's potential to drive significant advancements in biomedical research.

MAMMAL in the Cure Cancer With AI API

With the integration of MAMMAL into our API, researchers can now access three powerful endpoints that facilitate critical predictions in the drug discovery process. Each endpoint is accessible through a simple POST request and returns results in JSON format. Here’s a breakdown of what each endpoint offers:

• Protein–Protein Interaction (PPI) Prediction: - Endpoint: /api/v1/mammal/ppi - Input: Two amino-acid sequences (designated as protein_a and protein_b). - Output: A binding-affinity class label of "1" (indicating the proteins interact) or "0" (indicating no interaction).
• Drug–Target Interaction (DTI) Prediction: - Endpoint: /api/v1/mammal/dti - Input: A target protein amino-acid sequence (target_seq) and a drug represented in SMILES notation (drug_seq). - Output: A predicted pKd value (−log₁₀ Kd), where a higher value indicates a stronger predicted binding affinity between the drug and the target protein.
• ClinTox Clinical-Trial Toxicity Prediction: - Endpoint: /api/v1/mammal/clintox - Input: A compound represented in SMILES notation (smiles). - Output: A toxicity prediction, where "1" indicates the compound is toxic or likely to fail trials, and "0" indicates it is not toxic. Additionally, a raw score is provided to give further insight into the prediction.

These MAMMAL endpoints are now available for free at curecancerwithai.com. This API functionality sits alongside our existing free data API, which includes research papers, news, blog posts, FDA approvals, clinical trials, and cross-dataset search capabilities.

Start Using It for Free

Getting started with MAMMAL is easy! Anyone can create a free API key at /api-keys. With your API key, you can make up to 100 requests per hour, allowing you to explore the potential of MAMMAL without any cost. For full documentation, parameters, and code samples, visit /developers. Remember, the API calls must be authenticated with an "Authorization: Bearer ccw_live_..." header, and the key must remain server-side for security.

Conclusion

We are thrilled to bring the capabilities of the IBM MAMMAL model to the research community through our free public API at curecancerwithai.com. By providing access to these advanced predictions, we hope to empower researchers in their quest to understand cancer and develop new therapies. Please keep in mind that the predictions made through this API are research signals and should not be considered medical or clinical advice.

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.