The non-coding, intergenic component of the human genome is receiving increased attention from biologists. Finding functional elements within this >84% of the human genome presents major intellectual and experimental challenges. By comparing genomic DNA sequences from diverse species, functional elements may be recognized on the basis of their evolutionary conservation. This type of comparison has been performed on the major vertebrate genomes. We have extended this comparison to include all 168 genomes sequenced to date. By including organisms as diverse as Halobacterium, Agrobacterium tumefaciens, Wigglesworthia glossinidia, Escherichia coli, Ciona intestinalis, Arabidopsis thaliana, Caenorhabditis elegans, Anopheles gambiae, Drosophila melanogaster, Takifugu rubripes, Gallus gallus, Mus musculus, Canis familiaris, Pan troglodytes, and Homo sapiens we hoped to identify DNA sequences that, although non-coding, might have been essential for the evolution of life. An a-priori statistical analysis indicated that it was infinitely improbable that a sequence would be present in all 168 genomes. We used a motif-oriented machine learning method based on the Relevance Vector Machine algorithm and a modified hidden Markov model running on a cluster of twenty-one parallel dual-processor G5 computers to identify shared sequences. After 1008 hours of continuous computation, we identified the sequence, TGTATAAACGCACCAACAAATGAT. We repeated the analysis excluding the human genome. Surprisingly, this yielded two additional sequences, TATACAAGAATACATACGAGTGTTTTGCCAGTATTATGGACT and GTTTTATGGACAAGTGTATTGCCATATACGAGAATACATACT. The predicted amino acid sequences for these 3 highly conserved DNAs are CINAPTND, YTRIHTSVLPVLWT and VLWTSVLPYTRIHT respectively. The significance of these results is clear: during evolution within the genus homo, the meaning of life, the universe, and everything was lost.
(last revised 22 February 2005)
Return to
Zebrafish Homepage
|
