Combining large-scale experiments and modeling to dissect the effect of FIXvariants on secretion, carboxylation, activation and disease - Normal hemostasis critically depends on adequate circulating levels of coagulation factor IX (FIX). Lower FIX levels are associated with increased risk of bleeding and diagnosis with a bleeding disorder, while high FIX activity levels are associated with risk of thrombosis. FIX levels can be influenced by genetic variants in the F9 gene. Pathogenic F9 variants cause the bleeding disorder hemophilia B. Genotyping is clinically recommended for everyone at-risk to inherit or who has hemophilia B, where this information is useful for diagnosis, reproductive planning, and family testing in both disorders. In research, F9 sequence data inform studies of the importance of FIX for human health. Unfortunately, the majority of genetic variants detected during sequencing have insufficient evidence to classify with regard to pathogenicity. For example, in the My Life, Our Future (MLOF) hemophilia genotyping project, more than half of variants clinically suspected to cause hemophilia were classified as “Variants of Uncertain Significance” (VUS) by the clinical laboratory due to lack of evidence regarding pathogenicity. We recently developed a novel platform called MultiSTEP to interrogate the effects of FIX missense variation at scale. We applied MultiSTEP to generate variant effect scores for FIX secretion and carboxylation for all possible FIX missense. Our data revealed new insights into FIX biology and enabled re- interpretation of ~60% of missense VUS in hemophilia B. We hypothesize that the remaining unslved FIX VUS could be explained by variant effects on FIX functions other than secretion or carboxylation. This inability to accurately interpret the functional significance of genetic variants due to lack of evidence is a major obstacle to using the large volume of genetic information currently being produced in clinical sequencing and in efforts such as the NHLBI Trans-Omics for Precision Medicine (TOPMed) Program. In response to the current unmet needs and to the Notice of Special Interest, NOT-HL-23-066, we propose to measure the functional effect of nearly all possible F9 missense variants on FIX activation and inhibition and to develop models to predict disease severity and inform VUS reclassification. To accomplish our goals, we have adapted our system to display FIX on cultured human cells to measure the effect of thousands of missense variants simultaneously on FIX activation by FXIa in Aim 1, and FIXa inhibition by antithrombin In Aim 2,. In Aim 3, we will use the data generated in Aims 1 and 2 along with all other available data including our existing MultiSTEP secretion and carboxylation data to model variant impacts on FIX functions, disease severity, and classifications of pathogenicity. We will share our data and methods broadly, including via MaveDB, EAHAD, ClinGen and ClinVAR, and with investigators in TOPMed. Additionally, our variant effect scores will be used in MLOF to reclassify most, if not all, remaining VUS found in people with hemophilia B. Our work will reveal the relationship between F9 gene sequence and the mechanisms impacting FIX activities, as well as improving diagnostic accuracy for people with bleeding disorders.
