Numerous plants harbor viral sequences of known viruses in their genome albeit the integration in the host genome is not required in the life cycle of plant viruses. In banana, sequences of Banana streak virus (BSV) integrated within the B genome of banana plants as eBSV (endogenous BSV) spontaneously participate to plant infection by releasing a functional viral genome following stresses. The inability to identify such eBSV-risked plant has hampered the banana germplasm exchange and the use of the B genome in the breeding programs over the last 20 years. Besides, intensive cropping of banana containing risked eBSV such as plantain (cooking banana) represents a real threat of disease emergence not only for plantain but also for nearby plantations of dessert banana which is strongly susceptible to the disease. We elucidated the sequence and organization of eBSVs for three BSV species in the diploid M. balbisiana cultivar Pisang Klutuk Wulung (PKW), showing that eBSGFV and eBSOLV were di-allelic, with one infectious and one non-infectious allele, whereas that eBSIMV was monoallelic [1, 2]. Based on the sequences and the structure of these eBSV we developed several PCR and Derived Cleaved Amplified Polymorphic Sequences (dCAPS) markers and tested their robustness by genotyping each eBSV from numerous Musa samples covering a wide range of Musa diversity [3]. We used them to currently establish the eBSV ID of banana candidates (from the international Musa Germplasm Transit Center (ITC) (KU Leuven), EMBRAPA collection…) and their potential BSV phenotype. Our results and knowledge recently gained on eBSV allowed us to lift the international moratorium established in 2000 on the B genome movement, and its use in breeding programs. As part of the Musanet Conservation Thematic Group, which is composed of an international panel of scientific experts, we recently established a strategy (decision tree) for the safe distribution of eBSV-containing Musa germplam through Quarantine