VADR

VADR, short for Viral Annotation DefineR, is an open-source software pipeline developed by the National Center for Biotechnology Information (NCBI) to automate the annotation and quality control of viral genome sequences submitted to GenBank. The project aims to standardize gene models and feature annotations across viral genomes and to improve the reliability of publicly available sequence data.

The pipeline relies on curated reference models to annotate features such as coding sequences, untranslated regions, and noncoding RNAs, and to verify sequence integrity. It combines homology-based methods with structure-aware analyses to assess reading frames, start and stop codons, and other structural constraints. VADR produces a GenBank-compatible feature table along with a validation report that highlights potential issues or discrepancies in the submitted sequence.

In practice, users submit viral genome sequences in FASTA format, and VADR runs a sequence of analyses to assign annotations and detect anomalies. Sequences that pass quality checks proceed to annotation export, while those with problematic regions generate detailed reports to guide review or re-submission.

VADR is part of the GenBank submission workflow and is maintained by NCBI. It is distributed with documentation and libraries of reference models, and is designed for researchers working with viral genomes who require standardized, automation-friendly annotation and quality assurance.