Protocol to detect spatio-molecular profiles underlying neuroimaging features in the human cerebellum

Imaging transcriptomics offers opportunities to uncover the genetic profiles underlying brain imaging-derived phenotypes (IDPs) but lacks explanation from gene to IDPs. Here, we present a protocol for combining imaging transcriptomics with gene set variation analysis (GSVA) to detect spatio-molecular profiles underlying IDPs in the human cerebellum. We describe the steps for data preparation, model training, model evaluation, key gene identification, and GSVA. Our protocol broadens the way to interpret the biological pathways shaping a wide range of neuroimaging-derived cerebellar properties. For complete details on the use and execution of this protocol, please refer to Wang et al.1.