Modern biological models based on stress-diathesis consider suicide an independent form of behavior, which is the basis for the search for specific genetic markers. The current review embraces all presently existing GWAS studies of suicidal phenotypes (suicidal ideation, attempts, and completed suicides). Fifteen studies based on this strategy and using different approaches to data analysis, including comparisons with existing databases on psychiatric genetics, pathway analysis, protein-protein interactions, functional grouping of genes, and enrichment analysis are discussed. In contrast to the candidate gene studies which intensively discussed associations with neurotransmitter systems of the brain (indol- and catecholamines, GABA, excitatory aminoacids, etc.), GWAS suggest a rather different set of factors. Many findings are among the genes involved in neurodevelopment, neuroplasticity, cell adhesion and migration, proliferation, and intracellular signaling, as well as immune system functioning. We consider that this confirms the relevance of the stress-vulnerability models implying a key role of the early development, which affects neuroplasticity. Stress as a systemic reaction of the organism demonstrates a significant role in the genesis of suicidal behavior. It should be noted that findings of different studies rarely coincide, thus demonstrating heterogeneous findings. This may be due to differences in bioinformatics approaches, description of phenotypes, and design of the study. Future accumulation of data especially considering the ethnic factor, increase in the sample size, and meta-analyses may clarify the polygenic nature of suicidal behavior and identify genetic markers that are valuable for both understanding the pathogenesis of suicidality and suicide prediction and prevention.