Array CGH analysis of genomic alterations in breast cancer sub-types
Below are links to supplementary information that accompanies array CGH work to characterize genomic alterations in breast cancer sub-types published in Loo et al. 2004 Cancer Research 64, 8541-8549.
This work was performed in collaboration with the Trask Lab.Supplementary information
- Map positions, placement and chromosome band information on the April 2003 genome assembly for the 4153 BAC clones used in the array CGH analysis of breast cancer tumors as a comma delimited text file or a downloadable MS Excel spread sheet. A sequence "domain" for each clone, defining the possible extent of any deletion/duplication is also listed in this table (see Clone Domains below).
The base pair coordinates on the genome assembly of the array clones were determined using either their full or partial accession (39% of clones), BAC end sequence pairs (34%) or using an STS marker and/or single BAC end sequence (23%). 22 of the subtelomeric clones have estimated positions that place them at the ends of the chromosomes. Clones placed by a single BAC end sequence only have their opposite end coordinate estimated to be 150 kb away. The placement method is listed for each clone in the above table.
- Figure showing the distribution of the 4153 clones used for the array CGH across the human genome. Each vertical blue line represents one BAC clone; the gray blocks represent heterochromatin and centromeres.
- Large scale view of the array CGH profile of a breast tumor. The normalized log2 ratios of the tumor and reference values for each of the 4153 BACs are plotted against the BAC genomic position. A smaller version of this figure is numbered Figure 4 in the submitted manuscript.
A deletion or duplication seen in a clone can extend beyond the coordinates given for the clone in either direction. The maximum extent of any copy number change can be approximately determined by the next unaffected clone on the array in either direction. This sequence "domain" for each clone is the smallest region that should be initially investigated for potentially interesting sequence features (e.g. genes) affected by the deletion/duplication.
For each clone on our array we have defined its domain as being from the end position of the previous non-overlapping BAC to the start of the next non-overlapping BAC. The deletion/duplication may be smaller than this domain, but may also be slightly larger than the domain, extending partially into the region covered by a neighboring BAC. This neighboring BAC may not look deleted/duplicated because the majority of the sequence it contains has not been deleted/duplicated in the test sample. With overlapping clones, the domains have been extended to the next non-overlapping BAC to ensure that domains are not underestimated in cases where the clone coordinates have been determined by only a short piece of sequence (such as STS's or partial accessions). In these situations the clone coordinates may not reflect the true ordering of these clones along the chromosome.
The domain coordinates provided are intended to be only a starting point for further examination of the clones and genes in the regions of interest, particularly in regions of overlapping BACs.Using the clone domain information to further investigate a region
The genome region containing a clone of interest can be viewed by entering either the clone name, or its domain coordinates into a genome browser. Commonly used genome browsers include:
If the clone name is searched for in a genome browser, the zoom function can be used to expand the region shown until the neighboring array BACs are seen. In this way the "domain" for the clone of interest will be seen. The coordinates given for the clone domains are for the April 2003 genome assembly. These coordinates can be entered directly into UCSC's genome browser if this genome assembly is picked for view.