Sequencing on Compact Disc?; Microgenomics of Breast Cancer; Better Binding Site Prediction

PATENT WATCH | Sequencing on Compact Disc? Burstein Technologies of Irvine, Calif., has entered the alternative sequencing strategy arena (see related article, Beyond Sanger: Toward the $1,000 Genome), winning patent protection for its hybridization-based sequencing approach, which employs the company's BioCompact Disc (BCDT) technology. (US patent 6,566,069, issued May 30, 2003) "Oligonucleotide arrays hold great promise in gene sequencing," inventor Jorma Virtanen writes in the application. But applying these to de novo sequencing projects is problematic, in that the number and length of required oligos becomes prohibitive. Burstein's approach offers a technical workaround. One embodiment first determines all 16-mer sequences in the sample, followed by all 8-mer sequences that flank unknown 11-mers. The former dataset fleshes out the latter, allowing "almost unequivocal deduction of the original sequence," Virtanen writes. "Only some long repeat sequences are outside the capability of this method," and these can usually be deduced using custom-made arrays. The patent describes four ways to generate the required collection of uniform-length sample oligonucleotides. It then offers a two-step method to pattern millions of target oligos onto a BCD. No other method--be it lithographic, ink-jet, or conventional printing--enables such high-density array fabrication in just two steps, Virtanen writes, adding that the approach also requires no sophisticated instrumentation. Officials at Burstein could not be reached for comment. --Ivan Oransky TECH BRIEF | Microgenomics of Breast Cancer Pathologists typically classify breast cancer by stage (disease progression) and grade (aggressiveness). Yet these metrics are imprecise treatment indicators, so Dennis Sgroi, director of breast pathology, Massachusetts General Hospital, and his colleagues at Harvard Medical School and Arcturus (www.arctur.com) of Mountain View, Calif., are pursuing a molecular alternative: expression profiling. The team coupled laser-capture microdissection, microarray analysis, and bioinformatics to obtain precise molecular fingerprints--catalogs of gene activity--of normal, precancerous, and cancerous breast epithelial cells from tissue biopsy specimens.1 The results of this approach, what Arcturus CEO Thomas Baer calls "microgenomics," were unexpected, says Sgroi. Stage phenotypes are highly divergent, and Sgroi says he "naïvely" anticipated concomitant global transcriptome differences. Instead, he found that most transcriptional variation occurs at the normal-to-precancerous transition, indicating that invasive potential could be transcriptionally apparent even in early precancerous lesions. The expression profiles did cluster strongly by tumor grade, suggesting that molecular fingerprinting could ultimately prove a more robust grading system than histopathology. Now, says Sgroi, the question, is whether molecular fingerprints can predict clinical outcome, paving the way for more accurate diagnoses and more personalized treatment regimens. Many patients would be spared unnecessary, toxic rounds of chemotherapy, he says, while others could be treated earlier and more aggressively. Reaping such benefits, though, will require further study. --Jeffrey M. Perkel 1. X.-M. Ma et al., "Gene expression profiles of human breast cancer progression," Proc Natl Acad Sci, 100:5974-9, May 13, 2003. SOFTWARE WATCH | Better Binding Site Prediction Many tools can confidently predict gene locations and boundaries, but annotating regulatory regions continues to bedevil researchers, says Wyeth Wasserman, Centre for Molecular Medicine and Therapeutics, University of British Columbia, Vancouver, because those regions are relatively short, highly degenerate, and often located far from the genes they control. One common solution, comparing sets of transcription factor binding site (TFBS) profiles to a single sequence, yields mostly spurious matches, so Wasserman (then at Sweden's Karolinska Institute) and his team developed ConSite, a free Web tool, to improve these predictions (www.jbiol.com/content/2/2/13). ConSite employs "phylogenetic footprinting," limiting its analysis to evolutionarily conserved regions. Users submit two orthologous sequences or a completed alignment, along with an optional cDNA sequence to limit searches to noncoding regions. The program aligns the sequences if necessary, checks conserved regions against a vetted list of over 100 TFBS profiles (users can also upload their own TFBS profiles), and outputs the results. This approach, says Wasserman, reduces false-positives 10-fold, with only a modest loss in sensitivity, so ConSite can theoretically guide experimental work. But there's still a lot of noise, he says. "That tells us that there's still more work to do." Planned improvements include the ability to align multiple sequences (another team recently analyzed phylogenetic footprints in six yeast strains1), look for combinations of TFBS, and provide quality control feedback. --Jeffrey M. Perkel 1. P. Cliften et al., "Finding functional features in Saccharomyces genomes by phylo-genetic footprinting," Sciencexpress, 10.1126/science.1084337, published online May 29, 2003. function sendData() { document.frm.pathName.value = location.pathname; result = false if (document.frm.score[0].checked) result = true; if (document.frm.score[1].checked) result = true; if (document.frm.score[2].checked) result = true; if (document.frm.score[3].checked) result = true; if (document.frm.score[4].checked) result = true; if (!result) alert("Please select") return result; } .myradio{background-color : #94bee5} Please indicate on a 1 - 5 scale how strongly you would recommend this article to your colleagues? Not recommended 1 2 3 4 5 Highly recommended Please register your vote

Sequencing on Compact Disc?; Microgenomics of Breast Cancer; Better Binding Site Prediction

Interested in reading more?

Become a Member of