BioMap

About BioMap

Develops AI foundation models, specifically the xTrimo Cross-Modal Transformer, trained on proprietary datasets of over 6 billion proteins and trillions of single-cell gene expressions to predict protein behavior and design novel molecules. These models enable the creation of therapeutic antibodies, industrial enzymes, and _de novo_ proteins with limited or no labeled data, addressing the challenge of data scarcity in life sciences. BioMap's platform also includes a leaderboard for evaluating task model performance, supporting researchers and industry partners in accelerating protein development.

```xml <problem> Traditional AI methods in life sciences require extensive labeled data for accurate predictions, but such data is often scarce, hindering the development of novel proteins and therapeutics. The complexity of protein behavior and multi-omic interactions further compounds the challenge of designing effective molecules with limited data. </problem> <solution> BioMap develops AI foundation models, specifically the xTrimo Cross-Modal Transformer, trained on a large, proprietary dataset of protein sequences, protein-protein interactions, and single-cell gene expression measurements. These models enable the creation of therapeutic antibodies, industrial enzymes, and _de novo_ proteins, even with limited or no labeled data. BioMap's platform includes task models and a leaderboard for evaluating their performance, supporting researchers and industry partners in accelerating protein development and discovery. The xTrimo model leverages a super-computing center and high-throughput wet labs to understand and predict the behavior of life at multiple scales of complexity. </solution> <features> - xTrimo Cross-Modal Transformer: A pre-trained large language model trained on over 6 billion proteins, 100 billion protein-protein interactions, and trillions of single-cell gene expression measurements. - Task Models: Downstream models informed by the foundation model, facilitating generative AI for various applications. - Life Science Leaderboard: A tool for evaluating the performance of various task models for life science applications. - ImmuBot®: Multifunctional immunotherapy antibodies developed using BioMap's platform. - Capability to design next-generation antibodies and biologics against undruggable targets. - Ability to create breakthrough therapeutic proteins for gene, cell, and other new therapies. - Design of enzymes and _de novo_ proteins with enhanced properties for industrial applications. </features> <target_audience> The primary audience includes researchers and industry partners in the life sciences, including those focused on therapeutic antibodies, industrial enzymes, and novel protein design. </target_audience> ```

What does BioMap do?

Develops AI foundation models, specifically the xTrimo Cross-Modal Transformer, trained on proprietary datasets of over 6 billion proteins and trillions of single-cell gene expressions to predict protein behavior and design novel molecules. These models enable the creation of therapeutic antibodies, industrial enzymes, and _de novo_ proteins with limited or no labeled data, addressing the challenge of data scarcity in life sciences. BioMap's platform also includes a leaderboard for evaluating task model performance, supporting researchers and industry partners in accelerating protein development.

Where is BioMap located?

BioMap is based in Palo Alto, United States.

How much funding has BioMap raised?

BioMap has raised 100000000.

Location
Palo Alto, United States
Funding
100000000
Employees
87 employees
Major Investors
Hong Kong Investment Corporation Limited

Find Investable Startups and Competitors

Search thousands of startups using natural language

BioMap

⚠️ AI-generated overview based on web search data – may contain errors, please verify information yourself! You can claim this account with your email domain to make edits.

Executive Summary

Develops AI foundation models, specifically the xTrimo Cross-Modal Transformer, trained on proprietary datasets of over 6 billion proteins and trillions of single-cell gene expressions to predict protein behavior and design novel molecules. These models enable the creation of therapeutic antibodies, industrial enzymes, and _de novo_ proteins with limited or no labeled data, addressing the challenge of data scarcity in life sciences. BioMap's platform also includes a leaderboard for evaluating task model performance, supporting researchers and industry partners in accelerating protein development.

biomap.com3K+
cb
Crunchbase
Palo Alto, United States

Funding

$

Estimated Funding

$100M+

Major Investors

Hong Kong Investment Corporation Limited

Team (75+)

No team information available.

Company Description

Problem

Traditional AI methods in life sciences require extensive labeled data for accurate predictions, but such data is often scarce, hindering the development of novel proteins and therapeutics. The complexity of protein behavior and multi-omic interactions further compounds the challenge of designing effective molecules with limited data.

Solution

BioMap develops AI foundation models, specifically the xTrimo Cross-Modal Transformer, trained on a large, proprietary dataset of protein sequences, protein-protein interactions, and single-cell gene expression measurements. These models enable the creation of therapeutic antibodies, industrial enzymes, and _de novo_ proteins, even with limited or no labeled data. BioMap's platform includes task models and a leaderboard for evaluating their performance, supporting researchers and industry partners in accelerating protein development and discovery. The xTrimo model leverages a super-computing center and high-throughput wet labs to understand and predict the behavior of life at multiple scales of complexity.

Features

xTrimo Cross-Modal Transformer: A pre-trained large language model trained on over 6 billion proteins, 100 billion protein-protein interactions, and trillions of single-cell gene expression measurements.

Task Models: Downstream models informed by the foundation model, facilitating generative AI for various applications.

Life Science Leaderboard: A tool for evaluating the performance of various task models for life science applications.

ImmuBot®: Multifunctional immunotherapy antibodies developed using BioMap's platform.

Capability to design next-generation antibodies and biologics against undruggable targets.

Ability to create breakthrough therapeutic proteins for gene, cell, and other new therapies.

Design of enzymes and _de novo_ proteins with enhanced properties for industrial applications.

Target Audience

The primary audience includes researchers and industry partners in the life sciences, including those focused on therapeutic antibodies, industrial enzymes, and novel protein design.

Want to add first party data to your startup here or get your entry removed? You can edit it yourself by logging in with your company domain.