variable to include in the models, and studies are underway in our lab to examine FC measures in high-risk individuals. Fourth, although eLORETA has been reported to have excellent localization accuracy as validated by several multimodal studies, caution needs to be exercised especially while considering the findings for any clinical applications; it should be kept in mind that the FC measures used in the study are the derivatives of current density sources which are in turn derived by an inverse solution employed in the eLORETA algorithm which, like other similar methods, depends on several model constraints [166]. Fifth, the RF model in the current study has not included genomic factors (e.g., specific molecular genetic variants, polygenic scores, etc.), and future studies may also explore connectome genetics [167], which is advancing on many fronts and promises to shed light on how disease risk genes affect the brain connectivity [168,169]. Sixth, although there were seemingly interesting correlations among several connections within and across frequencies (Figure 7), they were not discussed as it was beyond the scope of the current study. Seventh, although DMN is the most widely studied network, comparison across other resting state and task-related networks may further our understanding of
