We obtained the GSE125367 dataset, from the Gene Expression Omnibus (GEO), in which genome‐wide DNAm profiles were assessed in whole blood samples using the Illumina MethylationEPIC array (Chater‐Diehl et al., 2019 (link)). This dataset includes 17 cases with variations in SMARCA2 and 27 neurotypical controls. Of the 17 cases, 12 were clinically diagnosed as NCBRS. They assessed genome‐wide DNAm profiles of eight cases, who had pathogenic variations in SMARCA2 and a clinical diagnosis of NCBRS, and 23 controls, selected to be age and sex matched to cases, and generated a model to determine whether the variation in SMARCA2 is benign or pathogenic. This model was applied to the remaining nine cases: four cases with uncertain variants and a clinical diagnose of NCBRS, one case with uncertain variants but not a clinical diagnose of NCBRS, and four cases with benign variants but not a clinical diagnose of NCBRS. As a result, the four cases with uncertain variants and a clinical diagnose of NCBRS were determined to be pathogenic, and the one case with uncertain variants but not a clinical diagnose of NCBRS was determined to be marginal. All of the four patients who were not diagnosed with NCBRS were determined to be benign. Therefore, we analyzed 12 cases with a clinical diagnose of NCBRS and 27 controls.