Theoretical Population Biology 61, 423–434 (2002) doi:10.1006/tpbi.2002.1589 Critical Issues in Bacterial Phylogeny Radhey S. Gupta 1 and Emma Griffiths Department of Biochemistry, McMaster University, Hamilton, Ontario, Canada L8N 3Z5 Received April 10, 2002 To understand bacterial phylogeny, it is essential that the following two critical issues be resolved: (i) development of well-defined (molecular) criteria for identifying the main groups within Bacteria, and (ii) to understand how the different main groups are related to each other and how they branched off from a common ancestor. These issues are not resolved at present. We have recently described a new approach, based on shared conserved inserts and deletions (indels or signature sequences) found in various proteins, that provides a reliable means for understanding these issues. A large number of conserved indels that are shared by different groups of bacteria have been identified. Using these indels, and based simply on their presence or absence, all of the main groups within Bacteria can be defined in clear molecular terms and new species could be assigned to them with minimal ambiguity. The analysis of these indels also permits one to logically deduce that the various main bacterial groups have branched off from a common ancestor in the following order: Low G+C Gram-positive ) High G+C Gram-positive ) Clostridium–Fusobacteria– Thermotoga ) Deinococcus–Thermus–Green nonsulfur bacteria ) Cyanobacteria ) Spirochetes ) Chlamydia–Cytophaga–Bacteroides–Green sulfur bacteria ) Aquifex ) Proteobacteria 1 (e and d) ) Proteobacteria-2. ðaÞ ) Proteobacteria-3 ðbÞ and ) Proteobacteria-4 ðcÞ: The validity of this approach was tested using sequence data from bacterial genomes. By making use of 18 conserved indels, species from all 60 completed bacterial genomes were assigned to different groups. The observed distribution of these indels in different species was then compared with that predicted by the model. Of the 936 observations concerning the placement of these indels in various species, all except one were in accordance with the model. The placement of bacteria into different groups using this approach also showed excellent correlation with the 16S rRNA phylogenies with nearly all of the species assigned to the same groups by both methods. These results provide strong evidence that the genes containing these indels have not been affected by factors such as lateral gene transfers. However, such events are readily detected by this means and some examples are provided. The approach described here thus provides a reliable and internally consistent means for understanding various critical and long outstanding issues in bacterial phylogeny. & 2002 Elsevier Science (USA) INTRODUCTION An understanding of the evolutionary relationships among prokaryotic organisms constitutes a long cher- ished goal of microbiology and biological sciences (Kluyver and van Niel, 1936; Stanier, 1941; Stanier and van Niel, 1962). In view of their ancestral nature, an understanding of the evolutionary relationships among them is critical for understanding numerous fundamental questions such as the nature and origin of the first cell, origin of metabolisms, origin of photosynthesis and of the information transfer processes, and also the origin of the ancestral eukaryotic cell. There have been numerous attempts made in the past to group and classify prokaryotes based on resemblances in their morphological, biochemical and physiological characteristics (Orla-Jensen, 1909; Buchanan, 1925; Kluyver and van 423 0040-5809/02 $35.00 # 2002 Elsevier Science (USA) All rights reserved. 1 To whom correspondence should be addressed. E-mail: gupta@ mcmaster.ca.