On March 22nd, the team led by Hong Liang, a specially appointed professor at Shanghai Jiao Tong University, released the large model Venus, whose core capability is to enable proteins to achieve “directed evolution” of their functions. The team has combined AI with protein design and modification to establish the world’s largest protein dataset. Models trained on this dataset can accurately and efficiently predict and design protein functions, transforming protein production from “slow trial and error” to “efficient and precise design”. If you want to enhance a certain function of the target protein, you only need to provide the sequence information of the protein, and Venus can quickly provide a batch of protein sequence modification plans, and then combine experimental verification to obtain the optimal modification plan. It is reported that multiple proteins designed through the Venus series model have been industrialized and will assist in the diagnosis of diseases such as Alzheimer’s.
Scan code to share