How gaius-augustus/braker works
BRAKER is a command-line genome-annotation pipeline: users provide a genome and RNA-seq and/or protein evidence, and receive predicted gene annotations. The README describes automated training and prediction with GeneMark and AUGUSTUS, with extrinsic evidence informing predictions; it also describes BRAKER3's highly supported combined gene set. The graph includes these documented stages and substantial indexed scripts. Source excerpts do not show how the main Perl pipeline dispatches helpers, so those connections are omitted.
Main entry points: braker.pl BRAKER pipeline
Components 13
Annotation workflow
- BRAKER pipeline
scripts/braker.pl - Genome sequence
- RNA-seq / proteins
- Gene annotations
Gene prediction
- GeneMark training
- AUGUSTUS prediction
Sequence and evidence utilities
- GC content range
scripts/get_gc_content.py - ETP hint extraction
scripts/get_etp_hints.py - Training gene selection
scripts/ensure_n_training_genes.py - GeneMark sorting
scripts/sortGeneMark.py
Model refinement
- UTR decoration
scripts/stringtie2utr.py - BUSCO ID normalization
scripts/dev/rename_busco_genes.py
Other components
- Researcher
Connections 10
- Researcher → BRAKER pipeline: runs
README.md - Researcher → Genome sequence: provides
README.md - Researcher → RNA-seq / proteins: provides
README.md - BRAKER pipeline → GeneMark training: trains
README.md - BRAKER pipeline → AUGUSTUS prediction: trains and predicts
README.md - BRAKER pipeline → Gene annotations: produces
README.md - GeneMark training → Gene annotations: contributes predictions
README.md - AUGUSTUS prediction → Gene annotations: contributes predictions
README.md - RNA-seq / proteins → BRAKER pipeline: informs training and prediction
README.md - Genome sequence → BRAKER pipeline: input sequence
README.md