Integrating Single-Molecule Variant Phenotyping with Clinical Features Predicts Outcomes in KIF1A-Associated Neurological Disorder
Listed in
This article is not in any list yet, why not save it to one of your lists.Abstract
Rare monogenic diseases lack scalable prognostic frameworks, leaving patients and families without guidance after molecular diagnosis. Using KIF1A-associated neurological disorder (KAND) as a model system, we integrated single-molecule biophysical phenotyping of 91 pathogenic variants with longitudinal clinical characterization of 343 patients. Vineland Adaptive Behavior Scales (VABS) Adaptive Behavior Composite (ABC) and Growth Scales identified two divergent trajectories emerging after age 10: a stable group and a declining group characterized by seizures, abnormal EEGs, and optic nerve atrophy. A framework combining biophysical parameters, computational pathogenicity scores, and clinical variables classified patients as stable versus declining with AUC 0.834 and predicted VABS ABC scores with R² = 0.484 under leave-one-out cross-validation. For newly identified KIF1A variants, this approach offers a prognostic tool and a stratification strategy for clinical trials. Because kinesin motor function is mechanistically conserved, the framework may generalize to other kinesinopathies and to monogenic disorders amenable to quantitative functional assays.
