Nevertheless, such screening does not always 
produce a conclusive answer, as in about 10% of individuals opting for the test a VUS is identified. The identification of a VUS in the BRCA genes is a major issue for geneticists, counsellors, oncologists and carriers. In contrast to the families carrying deleterious mutations, where relatives can be offered predictive testing, and carriers can benefit from riskreducing interventions and/or targeted surveillance, such management is not possible in families with a VUS, and surveillance can only be based upon the extent of the cancer family history. Therefore, it is of utmost importance to assess the clinical significance of each variant. Currently the pathogenicity of a VUS is widely assessed by multifactorial probability-based model analysis, and ideally in combination with functional assay data. However these tools are predictive and the in vivo behavior of the protein is not scrutinized. Although functional assays have been established, most are domain- and function- specific. Indeed, because BRCA1 is a multi-domain protein involved in a number of key cellular processes, the results of such assays can often be misleading and may even result in inaccurate cancer risk assessment, as they do not take into account the possible impact on the function of the BU 4061T Proteasome inhibitor entire protein. In order to overcome these limitations we have set up an in vitro cellular system where we can achieve high protein expression levels of fulllength BRCA1 following transient transfection. This approach can form the platform for setting up a wide range of functional assays which will enable the simultaneous interrogation of the known key functions of the intact BRCA1 protein. Bioinformatics analysis, using PolyPhen and Align-GVGD yielded contradictory predictions regarding the pathogenicity of the p.Ser36Tyr BRCA1 variant, identified in 4 Cypriot families and in 13 cases of sporadic breast cancer. This gave us the motivation to set up a cellular system for classifying this variant. In our system, we have successfully cloned the full-length coding sequence of BRCA1 into a mammalian expression vector with an epitope sequence on its N-terminus to allow detection of exogenously expressed protein. We demonstrated high protein expression levels of the full length BRCA1 in transiently transfected HEK23T cells. Similarly, we detected full-length protein for the p.Ser36Tyr variant and the known pathogenic variant p.Cys61Gly. However, the expression levels of both Afatinib variants were lower, compared to the wild type protein. This suggests that these mutations may either interfere with protein expression or that the resulting proteins are not as stable as wild type BRCA1. Given these observations then it is likely that the p.Ser36Tyr variant is clinically significant due to protein instability. The heterodimerization of BRCA1 with BARD1 is believed to confer stability to BRCA1 and their interaction is mediated through their RING domains. As the p.Ser36Tyr mutation falls within the RING domain of BRCA1 we investigated whether the mutation affects the interaction with BARD1. Coprecipitation analysis demonstrated that the interaction is not abolished, but the levels of co-precipitated BARD1 in S-phase synchronized cultured cells, were not as high as those precipitated by the wild type protein.