Sequence Design
Sequence design focuses on computationally creating optimal biological sequences (DNA, RNA, proteins) with desired properties, often for applications in medicine and biotechnology. Current research emphasizes the use of machine learning, particularly reinforcement learning, Bayesian optimization, and generative models like flow networks and transformers, to efficiently explore the vast sequence space and optimize for multiple objectives simultaneously. These advancements are improving the speed and efficiency of designing novel biomolecules for various applications, such as drug discovery, diagnostics, and gene editing, reducing reliance on expensive and time-consuming experimental trials.
Papers
Transfer Learning Bayesian Optimization to Design Competitor DNA Molecules for Use in Diagnostic Assays
Ruby Sedgwick, John P. Goertz, Molly M. Stevens, Ruth Misener, Mark van der Wilk
Molecule Design by Latent Prompt Transformer
Deqian Kong, Yuhao Huang, Jianwen Xie, Edouardo Honig, Ming Xu, Shuanghong Xue, Pei Lin, Sanping Zhou, Sheng Zhong, Nanning Zheng, Ying Nian Wu