VADR - Viral Annotation DefineR

Version 1.6.4; June 2024

https://github.com/ncbi/vadr.git

VADR is a suite of tools for classifying and analyzing sequences homologous to a set of reference models of viral genomes or gene families. It has been mainly tested for analysis of Norovirus, Dengue, and SARS-CoV-2 virus sequences in preparation for submission to the GenBank database.

The VADR v-annotate.pl script is used to classify a sequence, by determining which in a set of reference models it is most similar to, and then annotate that sequence based on that most similar model. Example usage of v-annotate.pl can be found here. Another VADR script, v-build.pl, is used to create the models from NCBI RefSeq sequences or from input multiple sequence alignments, potentially with secondary structure annotation. v-build.pl stores the RefSeq feature annotation in the model, and v-annotate.pl maps that annotation (e.g. CDS coordinates) onto the sequences it annotates.

VADR includes 205 prebuilt models of Flaviviridae and Caliciviridae viral RefSeq genomes, created with a process similar to the one described here. Example usage of v-build.pl can be found here. An advanced tutorial on building VADR models using RSV as an example can be found here. To use v-annotate.pl with viruses other than the default set of 205, see 'Available VADR models'. For instructions on using VADR for SARS-CoV-2 annotation see this page.

v-annotate.pl identifies unexpected or divergent attributes of the sequences it annotates (e.g. invalid or early stop codons in CDS features) and reports them to the user in the form of alerts. A subset of alerts are fatal and cause a sequence to fail. A sequence passes if zero fatal alerts are reported for it. VADR is used by GenBank staff to evaluate incoming sequence submissions of some viruses (currently Norovirus, Dengue virus, and SARS-CoV-2). Submitted Norovirus, Dengue virus and SARS-CoV-2 sequences that pass v-annotate.pl are accepted into GenBank.

The homology search and alignment components of VADR scripts, the most computationally expensive steps, are performed by the Infernal, HMMER, FASTA, MINIMAP2 and BLAST software packages, which are downloaded and installed with VADR installation.

---

SARS-CoV-2 annotation using VADR

The v-annotate.pl script includes some special options specifically developed for SARS-CoV-2 annotation that increase speed (-s and --glsearch options) and provide better annotation for sequences with stretches of Ns (-r option). See this page for more information on using VADR to annotate SARS-CoV-2 sequences.

---

Available VADR models

VADR installation includes a default set of Caliciviridae models including Norovirus virus. The installation also includes a set of Flaviviridae models including Dengue virus. You can download additional pre-built models to use to validate and annotate viruses, including SARS-CoV-2, RSV, or cox1 genes. Importantly, to use a set of models other than the default Caliciviridae set, you will need to use either the --mdir and --mkey options, or the the -m, -i, -x and possibly -n options as described here.

See this page for a list of all available models and additional information.

---

VADR documentation

• VADR installation instructions
  • Installation using vadr-install.sh
  • Setting environment variables
  • Verifying successful installation
  • Further information
• v-build.pl example usage and command-line options
  • v-build.pl example usage
  • v-build.pl command-line options
  • Building a VADR model library
  • How the VADR 1.0 model library was constructed
• v-annotate.pl example usage, command-line options and alert information
  • v-annotate.pl example usage
  • v-annotate.pl command-line options
  • Basic Information on v-annotate.pl alerts
  • Additional information on v-annotate.pl alerts
• Advanced tutorial: building an RSV model library
• Explanations and examples of v-annotate.pl detailed alert and error messages
  • Output fields with detailed alert and error messages
  • Explanation of sequence and model coordinate fields in .alt files
  • toy50 toy model used in examples of alert messages
  • Examples of different alert types and corresponding .alt output
  • Posterior probability annotation in VADR output Stockholm alignments
• VADR output file formats
  • VADR output files created by all VADR scripts
  • v-build.pl output files
  • v-annotate.pl output files
  • VADR coords coordinate string format
  • VADR sequence naming conventions
• Available VADR model files (github wiki)
• SARS-CoV-2 annotation (github wiki)
• Rfam-based structural annotation of a viral genome sequence for use with VADR (github wiki)
• Development notes and instructions (github wiki)

---

Contributors

• VADR includes contributions and input from current and former colleagues at NCBI, including:

  Rodney Brister

  Vince Calhoun

  Sergiy Gotvyanskyy

  Eneida Hatcher

  Sophia Hu

  Ilene Karsch-Mizrachi

  Rich McVeigh

  Susan Schafer

  Alejandro Schäffer

  Lara Shonkwiler

  Beverly Underwood

  Yuri Wolf

  Linda Yankie

---

Reference

• The recommended citation for using VADR for SARS-CoV-2 analysis: Eric P Nawrocki; Faster SARS-CoV-2 sequence validation and annotation for GenBank using VADR. NAR Genom Bioinform. 2023 Jan 20;5(1)::lqad002. (2023). https://doi.org/10.1093/nargab/lqad002

• The recommended citation for non-SARS-CoV-2 use of VADR is: Alejandro A Schäffer, Eneida L Hatcher, Linda Yankie, Lara Shonkwiler, J Rodney Brister, Ilene Karsch-Mizrachi, Eric P Nawrocki; VADR: validation and annotation of virus sequence submissions to GenBank. BMC Bioinformatics 21, 211 (2020). https://doi.org/10.1186/s12859-020-3537-3

---

Questions, comments or feature requests? Send a mail to eric.nawrocki@nih.gov.