Comparative Genomics

Sequence comparison is a powerful tool for analyzing gene function. While mutational analysis identifies alterations that interfere with gene function, comparison of non-mutant sequences from different organisms will identify those sequences that do or do not undergo genetic drift and are therefore retained by natural selection.

The mouse and human genomes, which on average have 86% and 88% sequence identity at nucleotide and protein level respectively, are too closely related to a common ancestor to reveal the regions that are necessarily conserved for functional regions. The most effective comparison is when the two genomes are not too similar and not too different. Such is the case with comparison between Fugu rubripes (the Japanese pufferfish) and human.

The value of Fugu sequencing

The Fugu genome contains about the same number of genes as the human genome. At the same time, orthologs from the pufferfish and human are different enough such that valuable structural and functional information can be extracted by simple comparison. Comparison between human and the pufferfish will reveal genes common to vertebrates and genes unique to mammals and fish. Between human and Fugu, intron/exon structures of most orthologues are conserved, and the coding exons but not the introns are conserved. Gene-prediction programs usually work much better on Fugu genomic sequence than on human genomic sequence, presumably because the Fugu genome is so much smaller.

The experience of our laboratory and others illustrates the value of comparison with Fugu sequences in analyzing the human genome. For example, analysis of the 4q35 region was beset by a dispersed gene structure and the presence of repeats. In this case, comparison of human and Fugu sequences was essential in clarifying the organization of the region. Protein sequence comparison between pufferfish and human can reveal information about conserved protein domains and motifs. Short DNA elements that are responsible for transcription regulation are also conserved between human and the pufferfish. This enables us to identify regulatory elements by comparison of non-coding regions of orthologous human and pufferfish genes.

Selected References, with an abstract
  • Bioessays 1998 Sep;20(9):741-9 Something fishy in the rat brain: molecular genetics of the hypothalamo-neurohypophysial system. Murphy D, Si-Hoe SL, Brenner S, Venkatesh B

  • Proc Natl Acad Sci U S A 1997 Nov 11;94(23):12462-6 Transgenic rats reveal functional conservation of regulatory controls between the Fugu isotocin and rat oxytocin genes. Venkatesh B, Si-Hoe SL, Murphy D, Brenner S

  • Nature 1993 Nov 18;366(6452):265-8 Characterization of the pufferfish (Fugu) genome as a compact model vertebrate genome. Brenner S, Elgar G, Sandford R, Macrae A, Venkatesh B, Aparicio S

    Abstract

    Cloning and sequencing techniques now allow us to characterize genes directly instead of having to deduce their properties from their effects. This new genetics reaches its apotheosis in the plan to obtain the complete DNA sequence of the human genome, but this is far beyond the capacity of present sequencing methods. Small 'model' genomes, 'such as those of Escherichia coli (4.7 megabases (Mb) and yeast (14 Mb), or even those of Caenorhabditis elegans (100 Mb) and Drosophila (165 Mb), are better scaled to existing technology. The yeast genome will contain genes with functions common to all eukaryotic cells, and those of simple multicellular organisms may throw light on the genetic specification of more complex functions. However, vertebrates differ in their morphology and development, so the ideal model would be a vertebrate genome of minimum size and complexity but with maximum homology to the human genome. Here we report the characterization of the small genome (400 Mb) of the tetraodontoid fish, Fugu rubripes. A random sequencing approach supported by gene probing shows that the haploid genome contains 400 Mb of DNA, of which more that 90% is unique. This genome is 7.5 times smaller than the human genome and because it has a similar gene repertoire it is the best model genome for the discovery of human genes.