Figure 5

Cross species KC-SMART analyses: overlapping recurrent CNAs of human and mouse breast tumors. We analyzed aCGH data from mouse and human tumor groups with the KC-SMART algorithm using a kernel width of 20 Mb. Kernel Smoothed Estimate (KSE) curves are shown for gains and losses separately. For each human-mouse comparison, the upper section shows the KSE curve of the mouse tumor group (Mm) and the lower section shows the human tumor group (Hs). Regions that are gained or lost significantly more often compared to random are depicted in red above or below the KSE curves. Genes that map to a syntenic region of significant gain or loss are plotted in red on the KSE curves and are connected with gray lines between the species. The KSEs of the aCGH data were scaled such that the significance threshold determined by KC-SMART analysis was set at 1 for gains and at -1 for losses. (a) Top panel: the gains on the mouse Brca1 ^Δ/Δ ;p53 ^Δ/Δtumor group linked to the gains of the human BRCA1 tumor group (bottom panel). (b) Top panel: the losses on the mouse Brca1 ^Δ/Δ ;p53 ^Δ/Δtumor group linked to the losses of the human BRCA1 tumor group (bottom panel). (c,d) Idem: mouse Brca2 ^Δ/Δ ;p53 ^Δ/Δand human BRCA2 tumor groups (e,f) Idem: mouse p53 ^Δ/Δand the human control tumor groups. The genomic locations of MYC, RB1, AURKA and ERBB2 genes are shown. Cancer related genes that map in the overlapping regions are shown in Figure 6 (reduced list) and Additional file 6 (complete list).