Engineering synthetic phosphorylation signaling networks in human cells

Rapid in silico directed evolution by a protein language model with EVOLVEpro [Science]

We developed a novel machine learning model to rapidly evolve protein in silico for higher activity. We demonstrated SOTA performance for EVOLVEpro across DMS benchamarks and used it evolve a highly active genome editing toolbox and T7 RNAP that are orders of magnitude higher fiedlity than current used wild-type.

Programmable RNA-guided DNA endonucleases are widespread in eukaryotes and their viruses [Science Advances]

We discovered the first eukaryotic RNA-guided nucleases that are homologs of bacterial TnpB and CRISPR/Cas12 nucleases. This work revealed novel evolutionary insights on RNA-guided systems in eukaryotes. In addition, we performed comprehensive biochemical characeterization on fanzor’s adaption in eukaryotes and engineered it for mammalian cell genome editing.

Programmable eukaryotic protein synthesis with RNA sensors by harnessing ADAR [Nature Biotechnology]

We engineered the first robust mammalian RNA-sensor based on ADAR called RADARS. The sensor can be reprogrammed to track any RNA species inside eukaryotic cells and allow conditional cargo expression based on the presence/expression of target mRNA(s). We demonstrate that the system can be readily integrated into AAV, lentivirus, and synthetic mRNA to selectively turn on an arbitrary protein of interest. We showcase the use of this system in cell specific killing, lineage tracing and in vivo recording for reprogrammable cell control.

Drag-and-drop genome insertion of large sequences without double-strand DNA cleavage using CRISPR-directed integrases