Fri, May 16, 2003: Scientific American features report by BSI founders on chain letters
WATERLOO, Ont. -- BSI founders Ming Li and Bin Ma's penchant for genome classification and their ensuing research in the area has led to a featured report in the June issue of Scientific American, see the article.
"It's all about our new theory of how to classify objects or sequences," said Li, a holder of the Canada Research Chair in Bioinformastics at the University of Waterloo. "The sequences can be anything from genomes, music, to internet documents, to chain letters, to computer programs, to languages."
The study, titled "Chain Letters and Evolutionary Histories," shows how a new algorithm can be useful in inferring the family tree of anything that evolves over time, including chain letters, biological genomes (genetic material of organisms), languages and even plagiarized papers.
Li and Ma (a winner of PREA award in 2003) co-authored the study with Charles H. Bennett of IBM, a renowned physicist who has invented Quantum Teleportation, who has collected the chain letters over a period of 15 years.
The featured report in Scientific American leads off with: "In our hands are 33 versions of a chain letter collected between 1980 and 1995, when photocopiers, but not e-mail, were in widespread use by the general public. These letters have passed from host to host, mutating and evolving. Like a gene, their average length is about 2,000 characters. Like genomes, chain letters undergo natural selection and sometimes parts even get transferred between coexisting 'species'." The authors add that while the letters are an intriguing social phenomenon, they provide a test bed for the algorithms used in molecular biology to infer the phylogeny (evolutionary history) of trees from the genomes of existing organisms. "We believe that if these algorithms are to be trusted, they should produce good results when applied to chain letters." They found that by using a new algorithm general enough to have wide applicability to such problems, it was possible to reconstruct the evolutionary history of the 33 chain letters. They also determined that standard methods do not work as well on the letters.
BSI is a bioinformatics software company that depends on novel ideas and sophisticated mathematical and algorithmic analysis, such as demonstrated in this Scientific American article, to provide bioinformatics solutions to the pharmaceutical and biotech industry. Its PatternHunter software uses beautiful proprietary technologies to speed up Blast by hundred times. Its PEAKS software is the world's only MS/MS data peptide de novo sequencing software accurate enough for the international proteomics efforts, when the protein is novo. Its soon-to-be-released software RAPTOR ranked number 1 in CAFASP3/CASP5 among the individual automatic protein 3D structure prediction programs.
