We are excited to announce that the powerful IBM MAMMAL (Molecular Aligned Multi-Modal Architecture and Language) model is now accessible for free through the Cure Cancer With AI public API. Developed by IBM Research, MAMMAL is a state-of-the-art multi-modal biomedical foundation model that has the potential to revolutionize cancer research and drug discovery.

What is MAMMAL and Why Does It Matter?

MAMMAL is an innovative biomedical foundation model that integrates multiple types of biological data, including proteins, small molecules, and gene/omics information, into a comprehensive framework. By leveraging a unified sequence representation, MAMMAL is designed to handle various tasks in biomedical research, making it a significant tool for advancing our understanding of cancer and developing new therapies.

Multi-modal models like MAMMAL are crucial in the field of drug discovery because they allow researchers to analyze and predict interactions between different biological entities in a holistic manner. This capability is particularly beneficial for cancer research, where understanding the complex interplay between proteins and drugs can lead to more effective treatment options and improved patient outcomes.

The Research Behind MAMMAL

The capabilities of MAMMAL are detailed in the research paper published in npj Drug Discovery (Nature). This paper showcases MAMMAL's evaluation across approximately 2 billion biological samples and its performance on 11 diverse drug discovery tasks. Remarkably, MAMMAL achieved new state-of-the-art results on 9 of these tasks, demonstrating its effectiveness and reliability in predicting biological interactions.

MAMMAL in the Cure Cancer With AI API

Cure Cancer With AI now offers three distinct MAMMAL API endpoints, each serving unique predictive capabilities:

• Protein–Protein Interaction Prediction (PPI):

  This endpoint helps researchers understand whether two proteins interact with each other. By sending two amino-acid sequences (protein_a, protein_b) through the API, users receive a binding-affinity class label of "1" (interacting) or "0" (non-interacting). This information is crucial for elucidating protein networks and interactions that play a role in diseases like cancer.

  Endpoint: POST /api/v1/mammal/ppi

• Drug–Target Interaction Prediction (DTI):

  This API endpoint predicts the interaction between a drug and its target protein. Users can input the target protein's amino-acid sequence (target_seq) and the drug's structure in SMILES notation (drug_seq). The output is a predicted pKd value (−log₁₀ Kd), where a higher value indicates stronger predicted binding. Such predictions are essential for optimizing drug design and enhancing therapeutic efficacy.

  Endpoint: POST /api/v1/mammal/dti

• ClinTox Clinical-Trial Toxicity Prediction:

  This feature assesses the potential toxicity of a compound, which is a critical aspect of drug development. By submitting a compound in SMILES notation (smiles), users receive a toxicity prediction, with a result of "1" indicating potential toxicity or likely failure in clinical trials, and "0" suggesting it is not toxic. This information aids researchers in screening compounds before clinical testing.

  Endpoint: POST /api/v1/mammal/clintox

Start Using It for Free

Getting started with the MAMMAL API is straightforward and completely free! Anyone interested in utilizing these powerful predictive capabilities can create a free API key at /api-keys. Additionally, the full documentation, including parameters and code samples, is available at /developers. The free tier allows up to 100 requests per hour per API key, making it easy for researchers to explore MAMMAL's predictions without any financial commitment.

Conclusion

We are thrilled to provide access to the IBM MAMMAL AI model through our public API, now available for free at Cure Cancer With AI. This powerful tool enhances research capabilities in understanding protein interactions, drug-target relationships, and clinical trial toxicity, all essential components in the fight against cancer. Remember, predictions made through the MAMMAL API are research signals and not to be considered medical advice or clinical outcomes.

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.