ALEXEY VERAKSA, Ph.D.
assistant professor
University of Massachusetts

Fred, California,

Won't all biological classification ultimately be based on genetics rather than the current patchwork system? Someone in a museum told me that there would always be a need to take structure into account. Why wouldn't genetics alone be definitive? Are people working on this now?

Alexey Veraksa
assistant professor,
Biology Department,
University of Massachusetts
(former HHMI predoctoral fellow)

For several centuries, biologists have wrestled with a bewildering diversity of living things by assigning them to hierarchical groups that form a basis of traditional biological classification. Foundations of the modern classification system were developed by the Swedish naturalist Carl Linnaeus in the 18th century. Biological classification was originally devised for the ease of identifying organisms and was based entirely on morphology.

Modern (natural) classification strives to determine evolutionary relationships between groups, centering on the Darwinian principle of common descent. It takes into account data from developmental biology that sometimes reveal deep commonalities not immediately obvious from comparisons of adult forms. More recently, natural classification has relied more and more on molecular systematics, using information from nucleotide sequences. As a result, many branches of the phylogenetic tree have been disassembled and reassembled into patterns based on molecular identification rather than on the previous system of taxonomy based on morphology or physiological characteristics. For example, it is now commonly accepted that Archaea represents a separate group, whereas it was previously grouped with Bacteria. More recently, two large protostome clades were introduced, Lophotrochozoa and Ecdysozoa, based on 18S ribosomal RNA (rRNA) sequences.

Will a classification system based on nucleotide sequences eventually replace the current "patchwork" system that takes into account morphological and other types of data? My opinion is that phylogeny will depend more and more on the sequence data, but there will always be a necessity for morphological analysis, because molecular systematics has its own limitations (see ref. 1).

Let me mention four types of problems encountered in molecular phylogenetic analysis:

1. Limitations of sampling. Unlike paleontological data, which are plentiful and go back millions of years, molecular systematics is largely limited to analyzing extant organisms (with the exception of shorter sequences that can indeed be obtained from fossils; see ref. 2). Many bacteria isolated from natural populations cannot be cultivated in the laboratory, which is a requirement for DNA sequencing. Despite significant progress in genome sequencing, we are still very far from covering a large enough sample of organisms for a satisfactory reconstruction of evolution. It still takes a considerable effort to produce a complete genome sequence for a higher organism. According to the December 10, 2005, update of the Genomes Online Database (GOLD), we now have 332 published complete genome sequences and 1,402 ongoing projects, a total of "only" 1,734 genomes (see ref. 3). This number, while representing a breathtaking advance in the power of genome sequencing, pales in comparison with hundreds of thousands of species identified by modern taxonomy. But even if the majority of these genomes were sequenced, the task of reconstructing a unique tree of evolution based on these data is enormous.

2. Molecular phylogenetics introduces certain assumptions in its analysis that are essential for building evolutionary trees. These assumptions, such as monophyletic (one-ancestor) origins of certain groups or uniform mutation rates, may not fully represent real processes. The further back we reach in our reconstructions of the tree of life, the more ambiguity and controversy we encounter.

3. Different types of data (for example, 18S rRNA, mitochondrial genomes, and specific gene sequences) often yield different tree structures. Which one is real? This problem may become less important as more complete genome sequences become available.

4. Horizontal gene transfer, mobile genetic elements, and complex genome rearrangements complicate reconstruction of evolutionary trees.

On the basis of these and other difficulties of molecular phylogenetics, I think that good old morphological and developmental evidence will continue to be an indispensable tool in systematics.

References

1. Rokas, A., Kruger, D., and Carroll, S.B. Animal evolution and the molecular signature of radiations compressed in time. Science 310: 1933-38, 2005.

2. Partial sequences from extinct organisms are being generated with an increasing rate. For more information, see the following link:

http://www.ncbi.nlm.nih.gov/Taxonomy/taxonomyhome.html/index.cgi?chapter=extinct

3. Information on the progress of genome sequencing can be found at these two sites:

Genomes online (GOLD): http://www.genomesonline.org/

Entrez genome: http://www.ncbi.nlm.nih.gov/entrez/query.fcgi?db=Genome

01/16/08 18:06