Misplaced Pages

This is an old revision of this page, as edited by Robiriondo (talk | contribs) at 18:15, 19 December 2024 (Added startup founded by world-renowned researcher Eric Xing.). The present address (URL) is a permanent link to this revision, which may differ significantly from the current revision.

GenBio AI (legal name: GenBio.AI, Inc.) is a biotechnology and artificial intelligence company headquartered in Palo Alto, California, with satellite offices in Paris and Abu Dhabi. The company focuses on the development of the world’s first AI-Driven Digital Organism (AIDO), an integrated system of multiscale foundation models designed to simulate, program, and predict biological outcomes at various scales, including DNA, RNA, proteins, cells, and evolutionary data.

History

GenBio AI was founded in 2024 by Eric Xing and Le Song, prominent researchers in machine learning and computational biology. The company’s launch coincided with the presentation of six peer-reviewed technical papers at the Conference on Neural Information Processing Systems (NeurIPS) detailing the technical framework behind AIDO.

Technology

GenBio AI’s flagship technology is the AI-Driven Digital Organism (AIDO), which integrates six foundational models that span multiple levels of biological complexity:

• AIDO-DNA: A 7-billion-parameter model trained on genomic data from 796 species, designed for genomic function and structure analysis.
• AIDO-RNA: A 1.6-billion-parameter model focused on RNA structure prediction, genetic regulation, and vaccine development.
• AIDO-Protein: A computationally efficient model for studying protein functions and interactions.
• AIDO-Single Cell: Pre-trained on a dataset of 50 million human cells, capable of analyzing entire transcriptomes.
• Protein Structure Model: Specializing in three-dimensional modeling of protein folding and structure-function relationships.
• Evolutionary Information Model: Providing insights into molecular evolution.

The models are interoperable, enabling a unified platform for simulating and programming biological processes across molecular, cellular, and systemic levels. AIDO is noted for its multitasking efficiency, capable of solving up to 300 tasks simultaneously.

Applications

GenBio AI’s technology addresses critical challenges in medicine and biotechnology:

• Drug Discovery: AIDO accelerates the identification of potential therapeutics by simulating millions of compounds and predicting their biological effects.
• Personalized Medicine: The platform supports the creation of digital patient twins to design customized treatment plans and reduce adverse drug reactions.
• Disease Understanding: AIDO provides tools to study systemic interactions, enabling researchers to explore conditions such as cancer and neurodegenerative diseases.

Research Contributions

The company has published six technical papers outlining the methodologies behind AIDO. These include advancements in sparse transformers, retrieval-augmented learning, and large-scale biological data integration. The research establishes AIDO as a new standard in biological modeling.

Leadership

• Eric Xing: Co-founder and Chief Scientist, a pioneer in AI and computational biology.
• Le Song: Co-founder and Chief Technology Officer, specializing in AI applications in biological systems.

The advisory board includes prominent scientists such as:

• Eran Segal: Department of Computer Science, Weizmann Institute of Science.
• Fabian Theis: Director of the Institute for Computational Biology at Helmholtz Munich.

Global Presence

GenBio AI operates globally with its headquarters in Palo Alto, and additional labs in Paris and Abu Dhabi. The team includes experts from institutions such as Carnegie Mellon University, Stanford University, MBZUAI, and the Weizmann Institute of Science.

External Links

• Official Website

See also

• Babelnet
• Deepl
• vidby
• OpenAI

References

1. GenBio AI. (2024). Official Website.

2. Toward AI-Driven Digital Organism: A System of Multiscale Foundation Models for Predicting, Simulating, and Programming Biology at All Levels. Retrieved from .

3. Accurate and General DNA Representations Emerge from Genome Foundation Models at Scale. Retrieved from .

4. A Large-Scale Foundation Model for RNA Function and Structure Prediction. Retrieved from .

5. Mixture of Experts Enable Efficient and Effective Protein Understanding and Design. Retrieved from .

6. Scaling Dense Representations for Single Cell with Transcriptome-Scale Context. Retrieved from .

7. Balancing Locality and Reconstruction in Protein Structure Tokenizer. Retrieved from .

8. Retrieval Augmented Protein Language Models for Protein Structure Prediction. Retrieved from .

• Artificial Intelligence software
• Companies based in Palo Alto
• Software companies of the United States
• Generative artificial intelligence companies
• Large language models