Skip to content

MetaGenome Gene Prediction

Introduction

A fundamental step in the analysis of environmental sequence information is the prediction of potential genes or open reading frames (ORFs) encoding the metabolic potential of individual cells and entire microbial communities. OmicsBox provides two tools for this purpose:

  • FragGeneScan: predicts intact and fragmented genes on short sequencing reads using a hidden Markov model. It can also be applied to incomplete assemblies or complete genomes and handles sequencing errors, with no need to train specific models for different datasets.
  • Prodigal: a fast, unsupervised tool for protein-coding gene prediction in prokaryotic genomes and metagenomes. It runs on finished genomes, draft genomes, and metagenomes without requiring training data.

In short, use FragGeneScan for plain reads and Prodigal for assembled contigs or scaffolds. Both tools output the predicted genes as nucleotide and amino acid multi FASTA files, together with a GFF file describing the gene features.

References

  • Trimble WL., Keegan KP., D'Souza M., Wilke A., Wilkening J., Gilbert J. and Meyer F. (2012). Short-read reading-frame predictors are not created equal: sequence error causes loss of signal. BMC bioinformatics, 13, 183.