Obsessive–compulsive symptoms are frequently observed in various psychiatric disorders, including obsessive–compulsive disorder, schizophrenia, depression, and anxiety. However, the underlying anatomical basis of these symptoms remains unclear. In this study, we aimed to investigate the mechanism of schizophrenia with obsessive–compulsive symptoms by using diffusion tensor imaging (DTI)-based structural brain connectivity analysis to assess the network differences between patients with obsessive–compulsive disorder (OCD), patients with schizophrenia showing obsessive–compulsive symptoms (SCH), schizophrenia patients with obsessive–compulsive symptoms due to clozapine (LDP), and healthy controls (CN). We included 21 patients with OCD, 20 patients with SCH, 12 patients with LDP, and 25 CN. All subjects underwent MRI scanning, and structural brain connections were estimated using diffusion tensor imaging for further analysis of brain connectivity. The topology and efficiency of the network and the characteristics of various brain regions were investigated. We assessed baseline YALE-BROWN OBSESSIVE COMPULSIVE SCALE (Y-BOCS), Positive and Negative Syndrome Scale (PANSS), and 24-item Hamilton Depression Scale (HAMD-24) scores. Our results showed significant differences among the SCH, OCD, and CN groups (p < 0.05) in the MRI-measured degree of the following nodes: the superior orbitofrontal gyrus (25Frontal_Med_Orb_L), lingual gyrus (47Lingual_L), postcentral gyrus (58Postcentral_R), and inferior temporal gyrus (90Temporal_Inf_R). Additionally, we found significant differences in the degree of the brain regions 02Precentral_R, 47Lingual_L, 58Postcentral_R, and 90Temporal_Inf_R between the CN, OCD, SCH, and LDP groups (p < 0.05). These findings suggest that alterations in the degree of nodes might be the mechanism behind obsessive–compulsive symptoms in schizophrenia.