Abstract The forebrain, midbrain, hindbrain, five facial prominences, and pituitary gland develop between wk 4-7 of gestation in humans. Genetic defects that disrupt these processes cause a spectrum of developmental disorders with life-long consequences that range in severity from holoprosencephaly (HPE) to septo-optic dysplasia (SOD) to pituitary hormone deficiency (congenital hypopituitarism, CH). HPE patients have variable defects in forebrain, eyes, and pituitary, and severe cases are embryonic lethal. The triad of features diagnostic of SOD include optic nerve hypoplasia, midline brain abnormalities, and CH. Patients diagnosed with CH, but not HPE or SOD, sometimes have features associated with those disorders, including vision, hearing, and/or brain anomalies. The genetic causes of these disorders are highly heterogeneous and overlapping. Prominent amongst the genetic causes are several genes that affect sonic hedgehog (SHH) signaling, including CDON, GLI2, GLI3, HHIP, SHH, SIX3, and TGIF1. We screened a cohort of ~ 200 unrelated probands with CH and various associated features and identified rare, likely pathogenic variants and variants of uncertain significance (VUS) in the transcription factors GLI2 and SIX3. We confirmed pathogenicity of several GLI2 and SIX3 variants using a SHH signaling sensor cell line assay and transient transfection assay, respectively. VUS are a major impediment to delivering on the promise of genetic testing for molecular diagnosis, and it is daunting for individual laboratories to establish the variety of functional testing assays necessary for genetically heterogenous disorders. We propose to create a catalog of the functional effects of all possible variants in GLI2 and SIX3 using multiplexed assays of variant effects (MAVEs). This approach scales up the assays we have already developed for testing one variant at a time so that thousands of variants can be tested simultaneously, yielding quantitative functional information that assigns variants as gain of This high throughput system addresses the problem of variant interpretation by providing comparable information about the phenotypic consequences of single nucleotide variants, which will improve the translation of genetic information into diagnosis. MAVEs have been applied to understand the function of diverse genes and types of pathogenicity, from splicing to amino acid substitution and from signaling pathways to transcription factors. function, tolerated, or loss of function. Completing these aims will further our knowledge of GLI2 and SIX3 structure and function in disease and set the stage for using MAVEs to generate catalogs of functional annotation for other genes in the SHH pathway that cause craniofacial defects.