Breast cancer is a highly heterogeneous disease, and intra-tumoral heterogeneity is believed to fuel tumour progression and drug resistance. Therefore, understanding clonal diversity is a prerequisite for designing future therapeutic strategies to prevent and stop the progression of the disease. In this project, we developed a new optical barcoding library called LeGOTeknik, enabling the detection of the fluorescent barcodes in single cell RNA sequencing. Breast cancer cells from multiple cell lines and PDXs were optically barcoded to track their fate after transplantation into the mammary fat pad of NSG mice. Using this approach, we studied the contribution of each clone in primary tumours and metastatic sites and characterised the hallmarks of the few clones able to survive, spread, and form brain macro metastases. The ability to link the transcriptomic profile, clonal origin and specific organ tropism of cancer cells allowed us to identify a unique gene signature observed in cells that can metastasise to the brain. To our surprise, this specific gene signature was detected in the primary tumour and could therefore serve as a useful predictive tool. Interestingly, these cells overexpressed genes from the neurogenesis pathway and, more generally, harboured a fascinating profile of brain mimicry.