 |
||||||||
| Program | ||||||||
| Abstract | ||||||||
|
|
||||||||
| Gene networks and pathways in schizophrenia | ||||||||
| Zhongming Zhao | ||||||||
| Virginia Commonwealth University, Richmond, VA | ||||||||
| Systems biology is emerging as a powerful approach for studying the causal mechanisms of complex diseases including schizophrenia. Schizophrenia is a complex and debilitating psychiatric disorder with a lifetime prevalence of ~1% in the world population. The past decade has witnessed hundreds of reports declaring or refuting claims that candidate genes chosen for their location under linkage peaks or because of their known physiological or pharmacological properties associated with schizophrenia. Identification of potential schizophrenia susceptibility genes is expected to accelerate because of many genome-wide association studies (GWAS) of large samples. So far, no gene or marker has been found to be certainly associated with schizophrenia. It now remains a great challenge for an investigator to decide how to select the data from various sources and weigh and rank the information so that the best candidates can be selected for further investigation (e.g. replication). In this study, we collected and compiled the available schizophrenia candidate genes including several sets of genome-wide linkage scans, more than 2000 association studies, gene expression data, and literature search, and then prioritized these candidate genes using a multidimensional evidence-based gene prioritization approach. Our extensive evaluation including using the unbiased GWAS markers suggested its utility in follow-up bioinformatics analysis.
We next selected the prioritized genes for network and pathway analysis. We found that schizophrenia candidate genes have an intermediate level of connectivity in the whole human protein-protein interaction (PPI) network. Our results demonstrate that SZGenes tend to have intermediate connectivity and intermediate efficiency with which a perturbation can spread throughout the network relative to essential genes and non-essential genes. We compared schizophrenia-specific subnetworks and cancer-specific subnetworks, both were extracted from the human interactome, and found that schizophrenia genes do not have a strong trend on interacting with each other or clustering compared to cancer genes. We also identified several small subnetworks which likely have major contribution to the risk of schizophrenia. Our results are helpful for better understanding the genetic mechanisms of schizophrenia |
||||||||
|
|
||||||||
|
|
||||||||
|
|
||||||||
|
||||||||
