Building multiplexed genomic tools for editing and interpreting human variation

Abstract

Advances in DNA sequencing over the past twenty years have led to an unprecedented expansion in our ability to read DNA sequences and find genetic variation in individuals. However, our ability to manipulate and interpret genetic variation has not nearly kept pace with our ability to identify variants. Despite a growing need for new genomic tools in clinical medicine, very few have been sufficiently developed and validated to be adopted in clinical settings. In particular, two tools have shown promise for clinical use– Clustered regularly interspaced short palindromic repeat (CRISPR)-Cas9 based gene editing and multiplexed assays of variant effect (MAVEs)–but remain limited in scope. In Chapter 1, I introduce each of these two methods, review how they work, how they are used in clinical settings, and discuss their current limitations. In Chapter 2, I describe a method I developed to expand the types of proteins that can be studied using MAVEs to include secreted and extracellular proteins via library-compatible mammalian surface display. I apply this method to coagulation factor IX (FIX), the primary genetic cause of hemophilia B, to quantify variant effects on FIX secretion, identify structural constraints for folding and secretion, and apply these data to FIX variation to determine mechanisms of disease. I then expand my study of FIX beyond secretion to post-translational modifications (PTMs) of FIX, including γ-carboxylation of its GLA domain and enzymatic cleavage by coagulation factor XIa. I find evidence for strong structural constraint in the FIX GLA domain that incompletely correlates to its level of γ-carboxylation and instead identifies distinct mechanisms for variant effects in the propeptide and GLA domains. In Chapter 3, I switch from functional characterization of genetic variants to controlled gene editing and repair using CRISPR-Cas9. I describe a new method to increase CRISPR-Cas9 based gene editing precision without compromising efficiency. By co-addition of catalytically inactive Cas9 guide RNAs (dRNAs) that bind to off-target editing sites, I significantly reduce Cas9 editing without the need for blocking mutations that alter adjacent sequences and improve editing efficiency for targeted variant knock-in. Finally, in Chapter 4, I outline future challenges for each of these technologies, suggest routes for further application with these tools, and comment on the future of clinically useful genomic tools.