classification and regression tree (cart) analyses of genomic signatures reveal sets of tetramers that discriminate temperature optima of archaea and bacteria分类和回归树(cart)分析基因签名揭示歧视套四聚物温度古生菌和细菌的最适条件.pdfVIP

Archaea 2, 159–167 © 2007 Heron Publishing—Victoria, Canada Classification and regression tree (CART) analyses of genomic signatures reveal sets of tetramers that discriminate temperature optima of archaea and bacteria 1 2 2,3 BETSEY DEXTER DYER, MICHAEL J. KAHN and MARK D. LEBLANC 1 Department of Biology, Wheaton College, Norton, MA 02766, USA 2 Department of Math and Computer Science, Wheaton College, Norton, MA 02766, USA 3 Corresponding author (mleblanc@) Received March 5, 2007; accepted June 8, 2007; published online July 17, 2007 Summary Classification and regression tree (CART) analy- 1990s when Samuel Karlin and others began to apply compu- sis was applied to genome-wide tetranucleotide frequencies tational analyses to a growing suite of completely sequenced (genomic signatures) of 195 archaea and bacteria. Although genomes (Karlin et al. 1994, Fertil et al. 2005, Paz et al. 2006, genomic signatures have typically been used to classify evolu- Vasilevskaya et al. 2006). tionary divergence, in this study, convergent evolution was the In 1992, with only partially sequenced bacterial, archaeal focus. Temperature optima for most of the organisms exam- and eukaryotic genomes, as well as some completed virus ined could be distinguished by CART analyses of tetranu- genomes, Karlin and colleagues explored the relative abun- cleotide frequencies. This suggests that pervasive (nonlinear) dances of di-, tri- and tetra-nucleotides (Burge et al. 1992). qualities of genomes may reflect certain environmental condi- Karlin et al. (1994) focused on dinucleotide relative abun- tions (such as temperature) in which those genomes ev