OpenCRISPR – gene editing meets AI

OpenCRISPR, credit Profluent

Natural evolution has had four billion years to experiment with living matter. From now on, artificial intelligence will also help expand the catalog of possible and desirable molecules. These so-called ‘language systems’ are no longer limited to producing text or images, as exemplified by ChatGPT or Midjourney. They can now be instructed and utilized to design new proteins, thereby potentially creating improved CRISPR scissors.

Continue reading

CRISPR meets machine learning

art-ball-shaped-blue-1047877

credit McKylan Mullins

If a donor template is not provided when CRISPR cuts the DNA, broken ends are fixed by natural repairing mechanisms in a way that is considered stochastic and heterogeneous. This makes template-free editing impractical beyond gene disruption, right? Wrong, according to a study published in Nature by Richard Sherwood and colleagues. Continue reading