Robotic platform AGENTEX speeds up testing of engineered genetic codes

In a study published in Nature, scientists unveiled AGENTEX, a fully automated pipeline that designs, assembles and tests synthetic genetic codes in cell-free extracts. The platform uses a novel tRNA-sequencing method (tSCAN) to quantify aminoacylation of thousands of engineered tRNAs, revealing that the conserved 3′ CCA end tolerates many mutations when native tRNA modifications are present. Leveraging this flexibility, the authors built compressed genetic codes that reassign up to three sense codons and incorporate a non-standard amino acid, all while preventing unintended decoding by wild-type ribosomes through targeted ribosomal mutations.

High-throughput robotics, custom Python scripts and mass-spectrometric verification enable rapid iteration of code designs. The work demonstrates that parallel, orthogonal translation systems can be prototyped safely in vitro, opening pathways for recoded genomes and novel biopolymer synthesis.