Genome editing has seen remarkable progress, with new approaches aiming to make the technology safer and more effective. Researchers at Mass General Brigham have developed an innovative machine learning algorithm called PAMmla, which can predict the properties of approximately 64 million different genome editing enzymes. This breakthrough is expected to significantly reduce off-target effects and improve the precision of gene editing applications.
According to Ben Kleinstiver, PhD, the corresponding author of the study, ‘Our study is a first step in dramatically expanding our repertoire of effective and safe CRISPR-Cas9 enzymes. We demonstrated the utility of these PAMmla-predicted enzymes to precisely edit disease-causing sequences in primary human cells and in mice.’ It highlights how combining protein engineering with machine learning can lead to more targeted and safe gene therapies.
The new technology addresses the traditional limitations of CRISPR-Cas9, such as unintended DNA modifications at non-target sites. By predicting the protospacer adjacent motifs (PAMs) crucial for enzyme targeting, the researchers identified novel Cas9 enzymes with enhanced specificity. The experiments showed promising results in human cells and mouse models, particularly in targeting retinitis pigmentosa, a genetic disorder affecting vision.
Rachel Silverstein, a PhD candidate and lead author of the study, emphasized the impact of their work: ‘The PAMmla model can now be used by researchers to predict customized enzymes onboard specific use cases, creating a toolbox of safe, precise Cas9 proteins for diverse applications.’ This development could accelerate the progress of gene therapy and personalized medicine by making enzyme selection more efficient and tailored to individual needs.
The study’s findings, published in Nature, mark an important advancement in the field of gene editing, paving the way for safer therapies and more precise genetic modifications. As Dr. Kleinstiver notes, ‘Building on these findings, we are excited to have these tools utilized by the community and also apply this framework to other properties and enzymes in the genome editing repertoire.’
