We created a visualization tool called Circos to facilitate the identification and analysis of similarities and differences arising from comparisons of genomes. Our tool is effective in displaying variation in genome structure and, generally, any other kind of positional relationships between genomic intervals. Such data are routinely produced by sequence alignments, hybridization arrays, genome mapping, and genotyping studies. Circos uses a circular ideogram layout to facilitate the display of relationships between pairs of positions by the use of ribbons, which encode the position, size, and orientation of related genomic elements. Circos is capable of displaying data as scatter, line, and histogram plots, heat maps, tiles, connectors, and text. Bitmap or vector images can be created from GFF-style data inputs and hierarchical configuration files, which can be easily generated by automated tools, making Circos suitable for rapid deployment in data analysis and reporting pipelines.[Supplemental material is available online at http://www.genome.org. Circos is licensed under GPL and available at http:// mkweb.bcgsc.ca/circos. An interactive online version of Circos designed to visualize tabular data is available at http:// mkweb.bcgsc.ca/circos/tableviewer.]The continuing advances in speed, quality, and affordability of whole-genome analysis, including genome sequencing, have transitioned the comparative genomics field from the realm of comparing reference sequence assemblies to comparing assemblies of individual genomes. Whereas interspecies analysis leverages information about one species to further the understanding of biological mechanisms in another, comparative methods are now used to discover differences between individuals and the extent to which these differences affect response to the environment, such as susceptibility to disease and responsiveness to therapy.Our growing ability to collect enormous amounts of sequence information to support such studies is arguably outpacing the rate at which we devise new methods to store, process, analyze, and visualize these data. Any new approaches in data modeling and analysis need to be accompanied with corresponding innovations in the visualization of these data. To mitigate the inherent difficulties in detecting, filtering, and classifying patterns within large data sets, we require instructive and clear visualizations that (1) adapt to the density and dynamic range of the data, (2) maintain complexity and detail in the data, and (3) scale well without sacrificing clarity and specificity.

We report that human secretory breast carcinoma (SBC), a rare subtype of infiltrating ductal carcinoma, expresses the ETV6-NTRK3 gene fusion previously cloned in pediatric mesenchymal cancers. This gene fusion encodes a chimeric tyrosine kinase with potent transforming activity in fibroblasts. ETV6-NTRK3 expression was confirmed in 12 (92%) of 13 SBC cases, but not in other ductal carcinomas. Retroviral transfer of ETV6-NTRK3 (EN) into murine mammary epithelial cells resulted in transformed cells that readily formed tumors in nude mice. Phenotypically, tumors produced glands and expressed epithelial antigens, confirming that EN transformation is compatible with epithelial differentiation. This represents a recurrent chromosomal rearrangement and expression of a dominantly acting oncogene as a primary event in human breast carcinoma.

The recurrent translocation t(8;16)(p11;p13) is a cytogenetic hallmark for the M4/M5 subtype of acute myeloid leukaemia. Here we identify the breakpoint-associated genes. Positional cloning on chromosome 16 implicates the CREB-binding protein (CBP), a transcriptional adaptor/coactivator protein. At the chromosome 8 breakpoint we identify a novel gene, MOZ, which encodes a 2,004-amino-acid protein characterized by two C4HC3 zinc fingers and a single C2HC zinc finger in conjunction with a putative acetyltransferase signature. In-frame MOZ-CBP fusion transcripts combine the MOZ finger motifs and putative acetyltransferase domain with a largely intact CBP. We suggest that MOZ may represent a chromatin-associated acetyltransferase, and raise the possibility that a dominant MOZ-CBP fusion protein could mediate leukaemogenesis via aberrant chromatin acetylation.

ibirol@bcgsc.ca.

BackgroundSubclassification of ovarian carcinomas can be used to guide treatment and determine prognosis. Germline and somatic mutations, loss of heterozygosity (LOH), and epigenetic events such as promoter hypermethylation can lead to decreased expression of BRCA1/2 in ovarian cancers. The mechanism of BRCA1/2 loss is a potential method of subclassifying high grade serous carcinomas.MethodsA consecutive series of 49 ovarian cancers was assessed for mutations status of BRCA1 and BRCA2, LOH at the BRCA1 and BRCA2 loci, methylation of the BRCA1 promoter, BRCA1, BRCA2, PTEN, and PIK3CA transcript levels, PIK3CA gene copy number, and BRCA1, p21, p53, and WT-1 immunohistochemistry.ResultsEighteen (37%) of the ovarian carcinomas had germline or somatic BRCA1 mutations, or epigenetic loss of BRCA1. All of these tumours were high-grade serous or undifferentiated type. None of the endometrioid (n = 5), clear cell (n = 4), or low grade serous (n = 2) carcinomas showed loss of BRCA1, whereas 47% of the 38 high-grade serous or undifferentiated carcinomas had loss of BRCA1. It was possible to distinguish high grade serous carcinomas with BRCA1 mutations from those with epigenetic BRCA1 loss: tumours with BRCA1 mutations typically had decreased PTEN mRNA levels while those with epigenetic loss of BRCA1 had copy number gain of PIK3CA. Overexpression of p53 with loss of p21 expression occurred significantly more frequently in high grade serous carcinomas with epigenetic loss of BRCA1, compared to high grade serous tumors without loss of BRCA1.ConclusionHigh grade serous carcinomas can be subclassified into three groups: BRCA1 loss (genetic), BRCA1 loss (epigenetic), and no BRCA1 loss. Tumors in these groups show distinct molecular alterations involving the PI3K/AKT and p53 pathways.