WNT4 is a secreted protein that plays a critical role in the regulation of cell fate and embryogenesis. Biallelic variants in WNT4 have been linked to SERKAL syndrome, an autosomal recessive disorder characterized by 46,XX sex reversal and dysgenesis of the kidneys, adrenals, and lungs. SERKAL syndrome has only been described in a single consanguineous kindred with four affected fetuses. Additional features seen in a subset of affected fetuses included ventricular septal defect (VSD), congenital diaphragmatic hernia (CDH), and orofacial clefting (OFC). To determine if these additional features were likely to be caused by WNT4 deficiency, we used machine learning to compare WNT4 to genes known to cause VSD, CDH, and OFC. When compared to all RefSeq genes, WNT4's rank annotation scores for these congenital anomalies were 94%, 99%, and 98.5%, respectively, indicating a high level of similarity. We subsequently identified a second consanguineous family with SERKAL syndrome in which an affected fetus had CDH and an affected child had OFC. We then demonstrated that a subset of Wnt4 null embryos have perimembranous VSDs, anterior and posterior sac CDH, and soft palate clefts. These findings suggest that WNT4 deficiency can cause VSD, CDH, and palatal anomalies in mice and humans with SERKAL syndrome. These studies also suggest that our machine learning approach can be used as a candidate gene prioritization tool, and that targeted mouse phenotyping can serve as a means of confirming the roles of candidate genes in mammalian development.