Genome profiling: a realistic solution for genotype-based identification of species.

Species identification is the basis of Biology and has been carried out based on phenotype. Although some genes, such as that for 16S rRNA, have been used for species confirmation, identification of species based only on genotype has never been done before, although recent whole genome sequencing studies have demonstrated it to be possible in principle. However, it is evidently unrealistic for routine experiments of species identification. This paper clarifies that a very limited amount of information derived from a genome sequence is sufficient for identifying the species. It also proves that Genome Profiling [Nishigaki, K., Amano, N., and Takasawa, T. (1991) Chem. Lett. 1097-1100], TGGE analysis of random PCR products, can not only fulfill such requirements, but also serve as a universal method to analyze species. Thus, this compact technology can be used in many fields of biology, especially in microbe-related disciplines such as microbial ecology and epidemiology where exact knowledge about all members of a population is essential but previously difficult to obtain. This is the first demonstration that genotype-based identification of species is possible using a simple and uniform protocol for all organisms.

Whole genome sequence-enabled prediction of sequences performed for random PCR products of Escherichia coli.

The sequence of an unknown PCR product generated by random (and conventional) PCR could be determined without sequencing when it is provided with the template DNA sequence. Theoretically, this was based on formerly established ideas which assert that the amount of random PCR product mainly depends on the stability of the primer-binding structures and that the dynamic solution structure of DNA is essentially governed by the Watson-Crick base pairing. However, it has not been clear whether this holds true for larger genomes of mega- to gigabase size, beside the lambda phage genome (of 50 kb) used previously, nor has it been ascertained to uniquely specify the sequence of a random PCR product. Here, we jointly use two computer programs together with experimental data from Genome Profiling (i.e. TGGE analysis of random PCR products). The first procedure carried out by a newly remodeled computer program (PCRAna-A1) was shown to be competent to calculate a set of random PCR products from Escherichia coli genome DNA (4.7 Mb). The other procedure performed with another program (Poland-H) played a critical role in determining the final candidate sequence by theoretically offering the initial melting temperature and the melting pattern of unspecified candidate sequences. The success attained here not only proved our method to be useful for sequence prediction but also confirmed the above-mentioned ideas as rational. We believe that this is the first case to computer-utilize a genome sequence as a whole.

Species-identification dots: a potent tool for developing genome microbiology.

Identification of species has long been done by phenotype-based methodologies. Recently, genotype-based species identification has been shown to be possible by way of Genome profiling, which is based on a temperature gradient gel electrophoresis (TGGE) analysis of random PCR products. However, the results, though sufficient in information, provided by genome profiling were complicated and difficult to deal with objectively. To cope with this, a technology of utilizing species identification dots (spiddos), which corresponds to structural transition points of DNAs, was introduced. Pattern similarity score (PaSS), derived from spiddos, was shown to be usable for quantitatively measuring the closeness between genomes. This was demonstrated with the experiments applied to the genomes of Escherichia coli O157:H7 (19 strains). The same genomes were also examined by sequencing and RFLP methods in order to compare the effectiveness of these three methods. As a result, the spiddos method was shown to give reasonable results and to be the most advantageous for measuring the closeness between species in general. This means that spiddos is pushing the heavy gate open for genome microbiology.