Parkinson’s disease (PD) is a chronic, progressive neurodegenerative disease and represents the most common disease of this type, after Alzheimer’s dementia. It is characterized by motor and nonmotor features and by a long prodromal stage that lasts many years. Genetic research has shown that PD is a complex and multisystem disorder. To capture the molecular complexity of this disease we used a complex network approach. We maximized the information entropy of the gene co-expression matrix betweenness to obtain a gene adjacency matrix; then we used a Fast Greedy algorithm to detect communities. Finally we applied Principal Component Analysis on the detected gene communities, with the ultimate purpose of discriminating between PD patients and healthy controls by means of a Random Forests classifier. We used a publicly available substantia nigra microarray dataset, GSE20163, from NCBI GEO database, containing gene expression profiles for 10 PD patients and 18 normal controls (NC). With this methodology we identified two gene communities that discriminated between the two groups with mean accuracy of 0.88 ± 0.03 and 0.84 ± 0.03, respectively, and validated our results on an independent microarray experiment. The two gene communities presented a considerable reduction in size, over 100 times, compared to the initial network and were stable within a range of tested parameters. Further research focusing on the restricted number of genes belonging to the selected communities may reveal essential mechanisms responsible for PD at a network level and could contribute to the discovery of new biomarkers for PD.© 2020 IOP Publishing Ltd.