中文BioGeometry has recently completed its Pre-A round of funding. This round of funding was led by Jiangmen Ventures and followed by Zhipu AI and SJ Jiacheng, with continued support from existing investor GaoRong Ventures. The new funds will be used to accelerate the application of generative AI models in biosolutions and to advance the development of proprietary products. The long-term vision of BioGeometry is to create a generative AI platform that drives the transformation towards a digital, intelligent and sustainable future.
Biosolutions: The Future of Industry
Biosolutions is poised to be a key driver of the “Fourth Industrial Revolution,” with its market size expected to reach tens of trillions of dollars This field has attracted global investments, spanning multiple industries such as biopharmaceuticals, food, chemicals, medical beauty, and materials. The core challenge is discovering high-value molecules (e.g. antibody and peptide drugs/vaccines) and optimizing existing microbial strains and biocatalytic enzymes to improve production efficiency and reduce costs. Rapid and accurate discovery of highly active protein macromolecules is essential in various bio-manufacturing sectors.
Unlike traditional manufacturing, which relies on petroleum-based chemical synthesis and causes environmental pollution and high costs, biosolutions uses microbial and biocatalytic technologies to synthesize products from renewable biomass such as corn and straw. The new method is efficient, environmentally friendly, and sustainable. It is predicted that over 60% of future products will be produced using biosolution methods, with the market value expected to soar to $30 trillion, potentially surpassing the digital economy.
Pioneering GenAI for Biosolutions
Generative AI technologies, such as ChatGPT and Sora, have rapidly advanced in recent years, ushering biosolutions into a new era of digitalization and intelligence. High-throughput gene synthesis, sequencing, and editing technologies are driving exponential growth in biological data, opening new opportunities for AI applications. In life sciences, Google DeepMind’s AlphaFold2 achieved revolutionary progress in protein structure prediction. This year, AlphaFold3, based on diffusion generative models, extends to predicting the structure of all biomolecules and their interactions, garnering widespread industry attention.
BioGeometry is at the forefront of creating a generative AI protein design platform that empowers the entire biosolution field. Its core team pioneered the application of diffusion models for molecular 3D structure generation in 2021, with their key paper, GeoDiff, being one of the top 50 most cited AI papers in 2022.
Recently, BioGeometry released GeoFlow, a state-of-the-art generative AI protein design model capable of both predicting antigen-antibody complex structures and designing novel antibodies. Unlike widely-used Transformer architectures in natural language processing, GeoFlow uses a geometric deep learning base model to better capture the atomic-level interactions in 3D space. It also employs the latest flow matching model, which has been shown to be more efficient and robust compared to diffusion generative models. GeoFlow’s performance in antigen-antibody complex structure prediction rivals AlphaFold3, making it the first generative macromolecule design model in the industry to reach this level.
Based on the powerful foundation model GeoFlow, BioGeometry developed GeoBiologics, an AI-driven one-stop platform for protein design, securing licensing agreements with several domestic and international pharmaceutical companies. The company has also formed strategic partnerships with various upstream and downstream enterprises to promote generative AI in biosolutions. These collaborations include developing AI-driven biosolution platforms with Alibaba Cloud, co-developing generative AI models for antibody design with Sanyou Bio and BioMeta, and advancing protein and antibody development with Xiangyao Bio and Yinjia Bio.
In the future, BioGeometry will continue to optimize its generative AI protein design platforms, empowering the biomanufacturing industry and rapidly advancing the development of proprietary products.
Commitment to Innovation and Growth
BioGeometry CEO Dr. Tang Jian expressed gratitude for the support from Jiangmen, Zhipu AI, Shengjing Jiacheng, and GaoRong Ventures. “After a year and a half of development, BioGeometry has established two fundamental platforms: a generative AI large model for protein design and a high-throughput protein wet lab validation. We have successfully validated multiple projects in antibody, peptide, vaccine, and enzyme design. Moving forward, we will continue to optimize our platform and accelerate the implementation of our proprietary products.”
Gao Xinxin, Founding Partner of Jiangmen Ventures, praised BioGeometry’s world-class technical strength and the team’s rich experience in biopharmaceutical and synthetic biology industries. “Under Dr. Tang Jian’s leadership, BioGeometry’s protein design platform has been applied to numerous projects and established collaborations with leading enterprises. We are confident in BioGeometry’s potential to become a world-class high-tech company in the AI biosolution field.”
Zhipu AI highlighted their commitment to supporting enterprises in the large model ecosystem, emphasizing BioGeometry’s deep expertise in protein molecule discovery and design. “We look forward to continued deep cooperation.”