Endogenous processes and exogenous agents cause constant DNA damage. DNA double-strand breaks are among the most serious types of damage. They are mainly repaired by homologous recombination, where the BRCA2 protein plays a dominant role. Heterozygous germline BRCA2 mutations predispose to breast, ovarian, pancreatic and other types of cancer. The presence of a pathogenic mutation in patients or their family members warrants close surveillance and prophylactic surgery. Apart from clearly pathogenic mutations, variants leading only to a single amino acid substitution are often identified. Since the influence of these variants on cancer risk is unknown, they represent a major clinical problem.The aim of this paper is to summarize the current possibilities of predicting pathogenicity of BRCA2 variants. In some cases, genetic methods are able to classify variants with high probability; however, their use is often limited by low frequency of the variants or inaccessibility of samples for mRNA isolation or DNA from family members. Alternatively, functional assays performed in various cellular models may be employed. Multiple functional tests and cellular models are presented and characterized, including their advantages and limitations. A new model of human syngeneic cell lines developed by the authors is presented, in which one BRCA2 allele is deleted and the variant is introduced into the other allele by homologous recombination. This model has the potential to evaluate function of variants without some of the unwanted effects of the other models. Currently, this model is being applied to variants identified in patients with hereditary cancer predisposition in the Masaryk Memorial Cancer Institute.Functional assays in cellular models including a new model of syngeneic cell lines described by the authors have a great potential in evaluating clinical importance of unclassified variants in the BRCA2 gene, especially in cases where genetic tests are not applicable.