Prefrontal cortical synaptoproteome profile combined with machine learning predict resilience towards chronic social isolation in rats
Ontology highlight
ABSTRACT: Chronic social isolation (CSIS), an animal model of depression, generates two stress-response phenotypes: CSIS-resilient and CSIS-susceptible. However, the molecular mechanism underlying resilience to CSIS remains unclear. We aimed to investigate the prefrontal cortical synaptoproteome profile between CSIS-resilient, CSIS-susceptible and control rats in order to identify predictive proteins for the resilience phenotype. Comparison of CSIS-resilient versus CSIS-susceptible rats revealed downregulated glycolysis intermediate Aldoc and upregulated Cltc, Cam2a, Syn and Fasn proteins involved in neuronal transmission, synaptic vesicle trafficking and fatty acid synthesis. Comparison of CSIS-resilient and control rats revealed downregulated mitochondrial proteins Atp5f1b and Cs, and upregulated Prkcg, Slc17a7, and Sv2a proteins involved in signal transduction and synaptic trafficking. These proteins make the animal groups linearly separable, and 100% validation accuracy is achieved by standard machine learning models. In addition, we identified the most significant proteins for discriminating CSIS-resilient vs. CSIS-susceptible or control rats by using Support Vector Machine with greedy forward search and Random Forest, resulting in four panels of discriminative proteins. Proteomics-data-driven machine learning algorithms can provide a platform for accessing predictive features and provide additional insight into the molecular mechanisms of synaptic neurotransmission involved in stress resilience.
INSTRUMENT(S): Q Exactive
ORGANISM(S): Rattus Norvegicus (rat)
TISSUE(S): Brain
SUBMITTER: Dragana Filipovic
LAB HEAD: Dragana Filipović
PROVIDER: PXD047159 | Pride | 2024-04-23
REPOSITORIES: pride
ACCESS DATA