High-dimensional Human Disease Data
Large-scale human disease datasets are the foundation of our AI drug discovery process. Our datasets include publicly available omics/real-world data and proprietary data that has been driven by internally funded, large-cohort studies. These patient-derived high-quality datasets are the prerequisites for drug discovery that will shed light on the underlying mechanisms of disease biology and treatment responses.
Biology at Scale
We built an internal high-throughput automation lab, which features our proprietary disease models, customized optical sensors, and more. We utilize our automation lab and highly integrated workflow to enhance the collection of high-quality and accurate data. This data is then used to catalyze the process of new lead discovery and target validation, while supporting quick iteration loops for our AI engine.
Cutting-edge AI Engine
BioMap’s AI antibody design platform integrates high throughput biological technologies with machine learning processes to design advanced, complex, and next-generation antibodies. We use unsupervised learning to train a deep sequence representation model on hundreds of millions of protein sequences that span the diversity of evolutionary models. We apply this to downstream tasks such as antibody structuring and biological activity prediction.
High-performance Biological Computing Platform
We rely on the powerful capabilities of our unique partners, such as Baidu Smart Cloud, to develop supercomputing clusters, high-performance protein computing chips, high-performance graph databases, high-level data security, and privacy computing for biological computing needs. The engine supports EB-level massive data storage and computing requirements, which creates super-large-scale protein pre-training models, graph neural network target analysis, predicts protein structures accurately, and performs complex simulations.