Go Back

Self-identification of protein coding region in bacterial genomes

Stephane Audic and Jean-Michel Claverie

Training set - based methods are ubiquitous in sequence analysis, and particularly prominent in the field of Gene Identification. For sometime now, we have argued about the danger of these methods being overly conservative, and precluding the discovery of atypical genes. We will present our first reasonably successful attempts at designing a truly "objective" gene identification program, without making use of a training set or similarity search information.

Go Back