HERVarium self-contained local distribution for local interactive exploration of human endogenous retroviruses

📦 HERVarium – self-contained local distribution

This Zenodo record provides a self-contained, runnable distribution of HERVarium, an interactive Dash-based web application for exploring a genome-wide atlas of human endogenous retroviruses (HERVs).

The archive contains all files required to run HERVarium locally, including the application code, precomputed annotation assets, reference files, and a reproducible conda environment specification. Users can run the application by downloading and unpacking this archive, creating the conda environment, and launching the app — no additional data downloads or manual assembly are required.

This distribution is intended to support reproducibility, peer review, and long-term accessibility during the preprint and initial release phase.

🧬 Overview of HERVarium

HERVarium enables interactive, locus-level exploration of multiple HERV annotation layers through an embedded IGV genome browser and searchable data tables, including:

• Internal HERV loci and protein-coding domain annotations, with domain class and HMM coverage summaries (GyDB / DFAM)

• LTR regulatory annotations, including transcription factor binding motif (TFBM) matches

• LTR structural architecture, including U3/R/U5 segmentation, PBS and PPT predictions, and promoter / polyadenylation signal annotations

• Integrated genome browser (IGV) with locally served annotation tracks and optional external tracks

All datasets are precomputed and optimized for fast local interaction using Parquet, JSON, BED, and bigBed formats.

📁 What is included in this record

This Zenodo archive (tar.xz) contains:

Application

• HERVarium Dash application (app.py)

• Reproducible conda environment specification (environment.yml)

Precomputed annotation data

• Aggregated and query-ready tables (.parquet, .json)

• Metadata files describing domains, LTRs, and U3/R/U5 features

Genome and annotation tracks

• Genome reference files required by the embedded IGV viewer (FASTA + index)

• Locally hosted BED / bigBed tracks for:

  • Internal HERV regions

  • Protein-coding domains

  • LTR elements

  • U3/R/U5 segments

  • PBS / PPT

  • Promoter and poly(A) signal annotations

  • LTR TFBMs

Static web assets

• Stylesheets

• Icons and logos used by the web interface

The directory structure inside the archive matches the layout expected by the application, enabling immediate execution after extraction.

▶️ How to run HERVarium locally (summary)

1. Download and extract this archive:

   tar -xvf hervarium.tar.xz cd HERVarium

2. Create and activate the conda environment:

   conda env create -f environment.yml conda activate hervarium

3. Launch the application:

   python app.py

4. Open a web browser at:

   http://127.0.0.1:8050

The application runs as a local web server and does not require an internet connection for core functionality.

🧑‍🔬 Intended use

This record is provided to:

• Enable transparent peer review during the preprint stage

• Ensure long-term, citable access to the exact data and assets used by HERVarium

• Allow readers to interactively explore HERV annotation layers locally

• Support reproducibility of figures and analyses presented in the associated manuscript

🔗 Related resources

The datasets integrated into HERVarium are also available as standalone Zenodo records:

• LTR regulatory atlas (U3–R–U5, motifs, PBS/PPT)
  https://doi.org/10.5281/zenodo.17602210

• Internal HERV domain annotation (ORFs + HMM profiles)
  https://doi.org/10.5281/zenodo.16318927

The development repository for HERVarium is hosted on GitHub:

• GitHub repository: https://github.com/funcgen/HERVarium

📖 Citation

If you use HERVarium or this distribution in your work, please cite:

• This Zenodo record

• The HERVarium GitHub repository

• The associated preprint / manuscript describing the resource: 

Regulatory Features and Functional Specialization of Human Endogenous Retroviral LTRs: A Genome-Wide Annotation and Analysis via HERVarium

Tomàs Montserrat-Ayuso, Aurora Pujol, Anna Esteve-Codina

bioRxiv 2026.02.17.706328; doi: https://doi.org/10.64898/2026.02.17.706328