Trifonov

Le Jeudi 23 Mars 2000 à 14h30

au LRI, Salle 101

Edward N. Trifonov

Department of Structural Biology, The Weizmann Institute of Science, Rehovot 76100, Israel

How much more the sequences tell multiple overlapping codes and how one may treat them

Résumé/Abstract : Common believe based on textbooks is that the only thing the nucleotide sequences are worth is the protein-coding sequences. The rest 95 percent of, say, human genome is considered as junk, parasite-like, and, perhaps, dispensable.

Chimpanzees can be proud of their protein-coding sequences, 99% identical to human. If they would reject their non-coding sequences as we are almost ready to do, we would have to include them in human race.

Clearly, the non-coding sequences make quite a difference, by carrying important messages. The number of various codes contained in the sequences is approaching 20 and soon will shoot over. Even more exciting for an information specialist is the fact that the codes (messages) overlap, so that any given letter of a given sequence serves simultaneously to several codes. Usual complexity criteria do not apply to such sequences.