Module 2: Genome Annotation

Genome annotation is the process of identifying and labeling functional elements in a genome assembly — including genes, repetitive elements, regulatory regions, and non-coding RNAs. A raw assembly is just a string of nucleotides; annotation transforms it into a biologically interpretable resource.

Types of Annotation

Type What it finds Examples
Structural Gene boundaries, exons, introns, UTRs BRAKER3, Helixer
Functional Gene function, protein domains BLAST, InterProScan
Repeat Transposable elements, repeat families RepeatMasker
Quality Annotation completeness BUSCO, OMArk

Learning Objectives

  • Mask repetitive elements before gene prediction
  • Predict gene structures using evidence-based and ab initio methods
  • Assess annotation completeness and quality
  • Visualize annotated features in a genome browser

Hands-on Tutorials

Hands-on 1 — Getting Started

Introduction to genome annotation tools and workflows in Galaxy.

Galaxy Training Network tutorials: Genome Annotation Tutorials


Hands-on 2 — Basic Gene Structure Annotation

Gene structure annotation for sequences not yet at contig/scaffold level.

Key tools used:

  • RepeatMasker — identify and soft-mask repetitive elements
  • RNA-STAR — align RNA-seq reads to the genome for splice site evidence
  • BRAKER3 — ab initio gene prediction with RNA-seq and protein evidence
  • GFFread — process and convert GFF/GTF annotation files

Hands-on 3 & 4 — Annotation Using Pipelines

Running complete annotation pipelines in Galaxy and assessing outputs.

Key tools used:

  • Helixer — deep learning–based gene structure prediction
  • BUSCO — benchmarking annotation completeness using conserved gene sets
  • OMArk — proteome quality assessment

Hands-on 4 — Visualization with WebIGV

Exploring annotated features interactively in the browser.

Key tools used:

  • JBrowse / WebIGV — view GFF annotation tracks alongside the assembly sequence

Key Concepts

GFF3 / GTF — Standard file formats for storing genome annotation. GFF3 (Generic Feature Format version 3) and GTF (Gene Transfer Format) describe genomic features with their chromosomal coordinates.

Evidence-based vs Ab Initio — Evidence-based annotation uses experimental data (RNA-seq, protein homology) to guide gene finding. Ab initio prediction relies solely on statistical models trained on known gene structures.

Repeat masking — Before gene prediction, repetitive sequences are masked (replaced with lowercase letters or Ns) to prevent them from confusing gene finders.

BUSCO score — Measures the percentage of expected conserved genes found in an assembly or annotation. A high complete BUSCO score indicates a high-quality, near-complete result.

Useful Resources