Ab initio gene finder


Ab initio gene finder search for term

Utilizes only the DNA sequence of the target genome in making its prediction for gene annotation. Gene preduction is done by first identifying all the possible boundary signals (termed signal sensing), linking these into syntactically valid gene parse (and identifying content regions through a content sensor), scoring all possible parses, and then selecting the highest-scoring parse as the final prediction. Two popular frameworks for evaluating gene parses are Markov models and random fields. Examples are GENSCAN, AUGUSTUS, GlimmerHMM and SNAP for GHMM-based gene finders, CRF for conditional random field – based gene finder. (Edwards, 2009)