Eukaryotic gene prediction using GeneMark.hmm-E and GeneMark-ES

Area of Science:

• Bioinformatics
• Genomics
• Computational Biology

Background:

• Accurate identification of protein-coding genes is crucial for understanding eukaryotic genome function.
• Existing gene-finding tools require manual parameterization, which can be complex and time-consuming.
• Novel genomic sequences necessitate reliable methods for gene annotation.

Purpose of the Study:

• To describe the usage of GeneMark.hmm-E and GeneMark-ES for eukaryotic gene prediction.
• To highlight the state-of-the-art accuracy of these tools across diverse eukaryotic genomes.
• To emphasize the automated parameterization feature of GeneMark-ES.

Main Methods:

• Utilizing GeneMark.hmm-E for gene prediction in eukaryotic genomic DNA.
• Employing GeneMark-ES, which features unsupervised training for automatic parameter estimation.
• Applying these tools for gene annotation in newly sequenced genomes.

Main Results:

• GeneMark.hmm-E and GeneMark-ES demonstrate high accuracy in identifying protein-coding genes.
• These tools are effective for gene annotation in fungal, plant, and animal genomes.
• GeneMark-ES successfully automates parameterization through iterative self-training.

Conclusions:

• GeneMark.hmm-E and GeneMark-ES are powerful and accurate bioinformatics tools for eukaryotic gene finding.
• The unsupervised training in GeneMark-ES simplifies the gene prediction process.
• These programs are valuable resources for genomic research and annotation projects.