omics.co.in logo omics.co.in

15. Eukaryotic Structural & Functional Genome Annotation

⚡ Pipeline 15 of 16  ·  🧬 Genome Annotation

Fully automated structural gene prediction combining RNA-seq spliced alignments and large-scale cross-species protein homology using GeneMark-ETP and AUGUSTUS, followed by InterProScan domain annotation.

🔍 Ctrl + K
Annotation • Eukaryotic Authority: BRAKER3 / GeneMark-ETP / AUGUSTUS
Production Blueprint v4.5

15. Eukaryotic Structural & Functional Genome Annotation

Fully automated structural gene prediction combining RNA-seq spliced alignments and large-scale cross-species protein homology using GeneMark-ETP and AUGUSTUS, followed by InterProScan domain annotation.

Metric / RequirementSpecificationTarget Configuration
Core Engine / FrameworkRepeatMasker + BRAKER3 + TSEBRA + InterProScanNextflow / Snakemake Compatible
Benchmark Runtime~8 – 16 hours for a 500 Mb genome32 CPU cores / 64 GB RAM
Standard Containerquay.io/biocontainers/braker3:3.0.8--hdfd78af_0Docker • Singularity • Conda
Input → DeliverableAssembly FASTA + RNA-seq BAM + Protein FASTA → GFF3, Protein AA FASTA, InterPro TSV
Workflow Chain: Masked FASTA + RNA-seq BAM + Protein OrthoDB → BRAKER3 → TSEBRA → InterProScan → eggNOG-mapper
⚡

Step 1: De Novo Repeat Model Identification & Soft-Masking

RepeatModeler2 • RepeatMasker

Identifies repetitive DNA elements, retrotransposons, and simple repeats, soft-masking them with lowercase bases.

# Build de novo repeat library
BuildDatabase -name genome_db assembly.fa
RepeatModeler -database genome_db -threads 32 -LTRStruct

# Soft-mask genome
RepeatMasker -lib genome_db-families.fa -xsmall -pa 16 assembly.fa
Key Flags & Notes: -xsmall soft-masks repeats (lowercase acgt) so gene prediction models can still read through.
⚡

Step 2: Integrated Gene Prediction with BRAKER3

BRAKER3 v3.0+

Trains GeneMark-ETP and AUGUSTUS using both RNA-seq split-read junction evidence and clade protein homolog sequences.

braker.pl --genome=assembly.fa.masked \
          --bam=rnaseq_spliced_aligned.bam \
          --prot_seq=/references/orthodb11_clade.fasta \
          --threads=32 --gff3 --useexisting
Key Flags & Notes: BRAKER3 achieves >90% sensitivity on complex eukaryotic genomes by integrating extrinsic RNA and protein evidence.
⚡

Step 3: Transcript Combiner & Support Filtering with TSEBRA

TSEBRA

Selects highest-scoring gene models supported by extrinsic RNA and protein alignment boundaries.

tsebra.py -g braker_output/braker.gtf \
          -e braker_output/hintsfile.gff \
          -o final_high_confidence_genes.gtf
Key Flags & Notes: Removes spurious un-supported ab initio predictions.
⚡

Step 4: Functional Domain Annotation with InterProScan

InterProScan v5.65+

Annotates protein families, Pfam domains, PANTHER hierarchies, and Gene Ontology (GO) terms.

interproscan.sh -i final_proteins.aa -f tsv,gff3 \
                -appl Pfam,PANTHER,Gene3D,SUPERFAMILY -goterms -pa -cpu 16
Key Flags & Notes: Yields complete GO term assignments and biological pathway classifications.

🚀 Automated Nextflow (nf-core) Production Execution

Execute this complete workflow across any SLURM, AWS Batch, or Google Cloud HPC cluster using the verified community container pipeline:

# Orchestrated via Singularity container or SLURM pipeline
← Return to Home Hub