Recognition of Genes in Human DNA Sequences
M.S. GELFAND, L.I. PODOLSKY, T.V. ASTAKHOVA and M.A. ROYTBERG
Journal of Computational Biology, 3(2):223-234 (1996)
Abstract
A new approach to computer-assisted gene recognition in higher eukaryote
DNA is suggested. It allows one to use not only linear functions for
scoring structures, but also functions satisfying natural monotonicity
conditions. The algorithm constructs the set of structures guaranteed
to contain an optimal structure for every function. So, it uncouples
the time-consuming step of generation of this set from the fast step of
structure scoring, thus making it simple to experiment with different
functions. One particular scoring function, taking into account only
codon usage and positional nucleotide frequencies of the splicing
sites, has been implemented in the Genome Recognition and Exon Assembly
Tool program, and has been tested on an independent sample of human
genes, yielding 88% sensitivity and 79% specificity.
