EGAPx (Eukaryotic Genome Annotation Pipeline)

The quality of EGAPx annotations is assessed using BUSCO scoring. BUSCO evaluates genome completeness by comparing the annotated gene set against conserved orthologous groups.

• High BUSCO score: Indicates a well-annotated genome with minimal missing or fragmented genes.
• Low BUSCO score: Suggests potential gaps or inaccuracies, or that the organism is highly divergent from the BUSCO models used.

EGAPx does not support organelle annotation. Because the pipeline is not aware of which sequences belong to organelles, it may produce inaccurate annotations for these sequences in the final GFF3 output.

Important:

• Do not use organelle annotations produced by EGAPx.
• The prepare_submission program will automatically delete organelle annotations when preparing files for GenBank submission.

Long RNA-seq reads can be provided alone or combined with short reads. They must be in FASTA or FASTQ format (not BAM). Use the long_reads: key in your YAML.

SRA Accessions or Entrez Query

long_reads: txid43150[Organism] AND (oxford_nanopore[Platform] OR pacbio_smrt[Platform]) AND biomol_rna[Properties]

Local Non-SRA Reads

Use a text file with unique set names and filepaths:

longset1 path/to/run1/long_reads.fq
longset2 path/to/run2/long_reads.fq

Full Configuration Example

genome: path/to/genome.fasta
taxid: 1234
short_reads: path/to/short_reads.txt
long_reads: path/to/long_reads.txt

genome: path/to/genome.fasta
taxid: 1234
short_reads: RNA-seq short reads data
long_reads: RNA-seq long reads data

EGAPx can be run in two modes depending on your network environment:

1. Online Mode: Support files are automatically staged from the internet during pipeline execution. This is the default behavior.
2. Offline Mode: Used for restricted internet access or reproducible runs. You must manually pull the Singularity image, download support files (via egapx.py), and provide a local cache directory using the -lc flag.

To test the pipeline, use the provided example: ./examples/input_D_farinae_small.yaml.

# Online mode run
python3 ui/egapx.py ./examples/input_D_farinae_small.yaml -e <executor> -w <workdir> -o <output>

# Offline mode run (requires -lc flag)
python3 ui/egapx.py edit_D_farinae_small.yaml -e biowulf_cluster -w <workdir> -o <output> -lc local_cache

After running prepare_submission, review the following files in your --out-dir before uploading to the NCBI Genome Submission Portal:

1. Submission ASN.1: Located at out_dir/annotated_genome.seq-submit.sqn.
2. Validation Output: out_dir/annotated_genome.seq-submit.val. Check for ERROR, REJECT, or FATAL issues. If you encounter unexpected issues with these labels, please open a GitHub issue.
3. Discrepancy Report: out_dir/annotated_genome.seq-submit.dr.
   • You can safely ignore FATAL errors named BACTERIAL_*.
   • The error Error: valid [SEQ_DESCR.BadStrucCommInvalidFieldName] Diploid is not a valid field is a false positive and can be ignored.

To install EGAPx, clone the repository, create a Python virtual environment, and install the required dependencies using pip.

Prerequisites

Before installation, ensure you have the following installed:

• Docker or Singularity
• AWS Batch, SLURM/UGE cluster, or an r6a.4xlarge machine (32 CPUs, 256GB RAM)
• Nextflow v.23.10.1
• Python v.3.11+

git clone https://github.com/ncbi/egapx.git
cd egapx

python3 -m venv venv
source venv/bin/activate
pip install -r requirements.txt

EGAPx uses MultiQC to generate a single HTML report summarizing the pipeline's performance and results. This report provides a centralized view of:

• Feature counts and length statistics: Summary of annotated features.
• BUSCO scores: Assessment of genome completeness.
• Genome masking statistics: Details on masked regions.
• Alignment statistics: Performance of RNA-seq and protein alignments.

Note: Current versions may lack data for protein alignments and local RNA-seq alignments. Check the Release Notes for status.

Before running the pipeline, you should run the command once to generate template configuration files in a ./egapx_config directory. This allows you to customize the environment settings.

Available Executors (-e flag):

• aws: AWS Batch using Docker image. Requires editing process.queue in ./egapx_config/aws.config.
• docker: Using Docker image (defaults to single node).
• singularity: Using Singularity image (defaults to single node).
• biowulf_cluster: Using Singularity image on Biowulf cluster. Requires editing ./egapx_config/slurm.config.
• slurm: Using SLURM in your HPC. Requires editing ./egapx_config/slurm.config.

Note: For local machine execution, no adjustments are needed.

# Generate config templates
python3 ui/egapx.py ./examples/input_D_farinae_small.yaml -e <executor> -w <workdir> -o <output>

# View help menu
python3 ui/egapx.py -h