The GeneMark-ES Gene Finding plugin offers an innovative approach to finding genes in eukaryotic genomes along with solving the parameter estimation problem by iterative self-training.
GENE PROBE Inc., the developers of the core components of GeneMark-ES, have developed an original implementation of the Viterbi algorithm for hidden semi-Markov Model of the eukaryotic genomic sequence. Gene Probe, Inc. has created and refined algorithms for gene prediction in genomic sequences for more than twenty-five years. GeneMark-ES gene finding plugin was further optimized for gene finding in fungal genomes.
GeneMark-ES is an ab initio gene finder designed to predict protein coding genes possibly interrupted with introns, a typical feature of gene organization in eukaryotic genomes. The algorithm parameters are genome-specific and are determined by unsupervised training on a given (novel) genomic sequence. The GeneMark.hmm algorithm uses a hidden semi-Markov model describing the functional and structural organization of a eukaryotic sequence.
Lomsadze A., et al. (2005) Gene identification in novel eukaryotic genomes by self-training algorithm. Nucleic Acids Research 33, 6494–6506.
Ter-Hovhannisyan V., et al. (2008) Gene prediction in novel fungal genomes using an ab initio algorithm with unsupervised training. Genome Research 18, 1979–1990.