Overview
Biogeometry is a generative AI-driven biotech company dedicated to "enabling generative AI to design the next generation of medicines." The company's core product, GeoBiologics, is a new generation AI platform for antibody discovery and protein design. The bottom layer is driven by the self-developed large-scale protein generation model GeoFlow V3 and a high-throughput wet experiment closed-loop platform, covering key aspects such as structural modeling, molecular optimization, de novo design, druggability prediction, and experimental verification. GeoBiologics can generate candidate antibodies, VHHs, small proteins or protein engineering plans based on target structure, target epitope and developability constraints, and continuously iterate the model based on wet experiment results, helping R&D teams shift from traditional "screening candidates" to "designing candidates". The platform can be used for de novo antibody design, antibody affinity optimization, antibody humanization, druggability prediction such as viscosity/expression level/stability/aggregation, as well as engineering of vaccine proteins, enzymes and other functional proteins. Compared with the traditional R&D process that relies on large-scale library screening and repeated trial and error, BIO Geometry improves early hit rates, shortens the discovery cycle, and reduces unnecessary experimental screening through the dry and wet closed loop of "computational design-experimental verification-model learning". The attached case shows that the platform has completed multiple VHH de novo design, IgG design, small protein design, antibody affinity optimization and vaccine protein engineering projects, and in some projects achieved a delivery rhythm of approximately 3 weeks to complete computational design and wet experiment verification.
Connected profiles
More Academic writing assistants products
Sources · 0Sources & provenance
English text is maintained separately from the original record. Consult the original source before citing a fact.
No source links are available for this profile.
Original source text
Names may use phonetic renderings when no English name is supplied. The original wording is retained below.
百奥几何是一家生成式 AI 驱动的生物科技公司，致力于“让生成式 AI 设计下一代药物”。公司核心产品 GeoBiologics 是面向抗体发现与蛋白设计的新一代 AI 平台，底层由自研蛋白质生成式大模型 GeoFlow V3 与高通量湿实验闭环平台共同驱动，覆盖结构建模、分子优化、从头设计、成药性预测和实验验证等关键环节。GeoBiologics 能够从靶点结构、目标表位和可开发性约束出发，生成候选抗体、VHH、小蛋白或蛋白工程方案，并结合湿实验结果持续迭代模型，帮助研发团队从传统“筛选候选物”转向“设计候选物”。平台可用于抗体从头设计、抗体亲和力优化、抗体人源化、黏度/表达量/稳定性/聚集性等成药性预测，以及疫苗蛋白、酶和其他功能蛋白的工程改造。相比依赖大规模文库筛选和反复试错的传统研发流程，百奥几何通过“计算设计-实验验证-模型学习”的干湿闭环，提升早期命中率，缩短发现周期，并减少不必要的实验筛选。附件案例显示，平台已完成多个 VHH 从头设计、IgG 设计、小蛋白设计、抗体亲和力优化和疫苗蛋白工程项目，并在部分项目中实现约 3 周完成计算设计与湿实验验证的交付节奏。