Ultrasome is an efficient methodology for detecting and delineating gains and losses of chromosomal material in DNA copy-number data. The methodology is originally described in:
Björn Nilsson; Mikael Johansson; Fatima Al-Shahrour; Anne E. Carpenter; and Benjamin L. Ebert. "Ultrasome: efficient aberration caller for copy number studies of ultra-high resolution". Bioinformatics (2009); doi: 10.1093/bioinformatics/btp091, in press.
Please cite this paper when using Ultrasome.
Microarray-based DNA copy number profiling is evolving rapidly in terms of genomic resolution. Recent microarrays, including Affymetrix SNP6.0, measure copy number at millions of chromosomal locations, an increase of up to ten-fold compared with earlier platforms. Even denser arrays are underway, and copy number profiling based on next-generation sequencing is rapidly gaining traction.
The central step in copy number data analysis is to partition the genome into contiguous regions that share the same copy number on average. With increasing resolution, this has become challenging as current standard methods, originally developed for lower-resolution microarrays, are associated with computational requirements that grow steeply with the number of probes. This leads to long computation times and complicates analysis.
To address this issue, we developed a new aberration caller, Ultrasome, that identifies changes-of-interest (amplifications and deletions) in near-linear time. Ultrasome processes latest-generation copy number arrays, including Affymetrix SNP6.0, about 10,000 times faster (less than one second) than standard methods (e.g., CBS) while retaining comparable analytic accuracy.
Download, license conditions
To download Ultrasome, click the link to the right. The program is available free-of-charge for academic use. The original work should be cited in publications that make use of the program.
Ultrasome is available in a command-line version (Windows and Linux) and a graphical user interface version (Windows, pictured here). Both versions offer the same functionality. The program accepts input data as tab-delimited text files (Broad Institute .cn format) and in a binary format (see User's manual, included in the download bundle). The program interfaces with Broad Integrative Genomics Viewer (http://www.broad.mit.edu/igv) to allow data visualization.
Further information, inquiries
The mathematical background, how to use, and how to install Ultrasome is described in Supplemental Methods and in the User's manual (both included in the download bundle). For specific inquiries and feature requests, contact bnilsson at broad dot mit dot edu.