Rice University researchers evolved compact CRISPR enzymes to access previously difficult PAM contexts, expanding editing capabilities for heart disease targets. Using an AI-supported platform to generate sequence-activity datasets, the team evolved SlugCas9 variants capable of engaging weakly bound PAMs. In cultured cell experiments, the evolved enzymes successfully edited disease-linked targets, including APOC3 and PCSK9, both associated with coronary heart disease risk. The strategy reframes PAM constraints as an engineering problem solved through rapid enzyme evolution rather than altering target DNA motifs. For gene therapy teams, the approach could widen the set of genomic sites compatible with compact CRISPR systems that are often preferred for delivery constraints.