<p>Gene expression analysis is an essential means for uncovering hidden biological knowledge and advance biomedical research. Gene co-expression network (GCN) analysis helps elucidate gene-gene relationships towards the identification of potential biomarkers. In this systematic and comprehensive survey, we present an in-depth review of the methodologies, techniques, and tools employed in GCN analysis, covering the entire pipeline from data preprocessing to network validation. We critically compare (diff) co-expression network analysis methods, co-expression measures, topological properties, network visualization and validation tools. Additionally, we highlight key challenges, current limitations, and unresolved issues in GCN research. We also highlight the increasing use of machine learning and deep learning methods, especially graph neural networks, which improve the ability of GCN analysis to handle large data, capture non-linear patterns, and model dynamic gene interactions. Finally, we offer practical recommendations on the selection of methods and tools for effective GCN analysis and propose future research directions to address the gaps in this rapidly evolving field.</p>