Skip to content

Repository files navigation

PrimeReplace Atlas

PrimeReplace Atlas maps ClinVar pathogenic/likely pathogenic variants onto MANE/GENCODE transcript structures to show how disease-associated variant burden is distributed across exons, exon blocks, and coding-payload units.

It provides a pre-implementation architecture layer for evaluating how public pathogenic variant records are distributed across transcript units relevant to large-fragment genome-writing hypotheses.

The atlas does not predict editing efficiency, safety, therapeutic efficacy, or patient coverage; it provides a pre-implementation target-architecture map.

Browser: https://best916116-crypto.github.io/PrimeReplace-Atlas/
Repository: https://github.com/best916116-crypto/PrimeReplace-Atlas
Zenodo archive: https://doi.org/10.5281/zenodo.20338345
Current release: v1.0.1, 2026-05-22
Release notes: reports/release_notes_v1.0.1.md
Data dictionary: reports/data_dictionary_v1.0.0.md
Validation summary: validation/public_release_checks_v1.0.1.json


Workflow schematic

PrimeReplace Atlas workflow schematic

The schematic summarizes the release logic: public ClinVar/MANE/GENCODE resources are mapped onto transcript units, converted into architecture labels and payload/context caveats, and exposed through browser pages and downloadable release tables. The claim boundary is explicit: PrimeReplace Atlas is a pre-implementation architecture resource, not an editing-efficiency, safety, patient-coverage, therapeutic-efficacy, or modality-recommendation system.


What it does

PrimeReplace Atlas converts public disease-associated variant records into transcript-level architecture summaries.

In practical terms, the atlas helps answer whether ClinVar P/LP records for a gene are:

  • concentrated in one exon,
  • distributed across an adjacent exon block,
  • grouped by a downstream coding-payload unit,
  • spread across a large or payload-burdened gene,
  • limited by sparse ClinVar record support,
  • affected by tumor-predisposition, repeat, noncoding, splice, isoform, or other mechanism-complex contexts.

The core output is an architecture map, not a yes/no feasibility call.


Why this is different

PrimeReplace Atlas does not replace ClinVar, MANE, GENCODE, genome browsers, or editing-design tools. Instead, it adds a transcript-unit architecture layer between public disease-variant records and downstream genome-writing design.

Resource type What it provides What PrimeReplace Atlas adds
ClinVar Pathogenic/likely pathogenic variant records Gene- and condition-level transcript-unit burden architecture
MANE / GENCODE Transcript anchors and exon/CDS structures Record-level coverage across exon, exon-block, and boundary-CDS units
Genome browsers Locus and transcript visualization Precomputed architecture labels and payload/context caveats
Editing-design tools Modality-specific design parameters Upstream pre-implementation architecture review

Most public variant resources answer what variants or annotations exist for a gene. PrimeReplace Atlas asks how disease-associated ClinVar P/LP records are organized relative to replacement-scale transcript units, and whether apparent record coverage is local, boundary-driven, payload-burdened, sparse, tumor-predisposition-associated, or mechanism-complex.

That derived architecture layer is the main novelty: it makes gene- and condition-level pathogenic-record organization reviewable before any experimental implementation, editing modality choice, construct design, or guide design.


Use it to ask

Use PrimeReplace Atlas to ask questions such as:

  • For a disease gene, where do ClinVar P/LP records fall across the MANE/GENCODE transcript?
  • Do records cluster in a local exon or adjacent exon block?
  • Does apparent high coverage depend on a large coding payload or boundary-driven CDS coverage?
  • Is the gene record-supported enough for architecture interpretation, or is it low-record-burden?
  • Are condition-associated ClinVar record groups available for disease-facing interpretation?
  • Which downloadable source tables support the gene or condition summary?

Open the browser

Start here:

https://best916116-crypto.github.io/PrimeReplace-Atlas/

The browser is freely available without login, registration, or password protection. It is a static public site, so the same release can be reviewed through GitHub Pages, the GitHub repository, or the archived Zenodo bundles.

Suggested first-use workflow:

  1. Open Gene search.
  2. Search a gene symbol, for example DMD, F9, ABCA4, CFTR, LDLR, or BRCA1.
  3. Open the gene page.
  4. Inspect the gene-level architecture label.
  5. Review unit-level record coverage and payload/context caveats.
  6. Open Condition facets to inspect condition-associated ClinVar P/LP record groups.
  7. Download source tables from the Downloads page.

What the atlas reports

PrimeReplace Atlas reports several independent interpretation axes.

Axis Meaning
Record support Number of ClinVar P/LP records available for a gene or condition-associated record group
Transcript-unit type Single exon, adjacent exon block, or coding-payload / boundary-driven CDS unit
Record-level coverage Fraction of ClinVar records covered by a transcript unit
Payload/context burden Donor span, coding-payload burden, boundary context, or related caveat
Architecture label Public-facing summary of how the pathogenic-record burden is organized
Condition facet ClinVar condition-associated record group, where available

The atlas deliberately keeps these axes separate. It does not reduce them to a single suitability score.


Architecture labels

PrimeReplace Atlas uses architecture labels to describe how pathogenic records are organized relative to transcript-unit replacement hypotheses.

Primary public-facing groups include:

  • Low-record-burden interpretation-limited: genes retained in the atlas but not overinterpreted because record support is sparse.
  • Local compact architecture: pathogenic records concentrate in a single exon or adjacent exon block.
  • Boundary-driven CDS coverage architecture: records downstream of a transcript boundary can be grouped by a coding-payload hypothesis; this is a burden-aware architecture descriptor, not a functional-rescue claim.
  • Donor/payload-burden caveat: high theoretical record coverage may require large or context-burdened payloads.
  • Tumor-predisposition control context: high-burden tumor-predisposition genes retained as control/caution contexts.
  • Mechanism-complex limitation: repeat, noncoding, splice, isoform, or mechanism-specific biology limits transcript-unit interpretation.

What this resource is not

PrimeReplace Atlas is not:

  • an editing-efficiency predictor,
  • a therapeutic-efficacy predictor,
  • a safety predictor,
  • a patient-coverage or prevalence estimator,
  • a final PA-family or genome-writing modality recommender,
  • a global ranking score,
  • a clinical decision engine.

PrimeReplace Atlas is a pre-implementation architecture resource for organizing public disease-associated ClinVar records relative to transcript-unit replacement hypotheses.


Data release

The current public release is v1.0.1. The biological table baseline is the v1.0.0 static-browser payload; v1.0.1 aligns citation and archival metadata for the deployed browser without changing the count denominators below.

Layer Count
ClinVar VCF records scanned 4,403,650
Primary ClinVar P/LP VCF records 337,682
ClinVar P/LP gene-variant rows 346,444
Unique genomic coordinates 336,901
All-mapped genes 5,738
Transcript-unit opportunity rows 437,301
Low-record-burden interpretation-limited genes 3,459
Architecture-interpretable genes 2,279
High-support condition-associated record groups 3,389
High-support condition groups with recovered unit coverage 1,278
Condition-unit coverage rows 160,125

DOI and version chain

PrimeReplace Atlas uses Zenodo DOI-versioned releases.

DOI type DOI Meaning
Concept DOI https://doi.org/10.5281/zenodo.20174921 Persistent DOI for the PrimeReplace Atlas release series
Current release DOI https://doi.org/10.5281/zenodo.20338345 v1.0.1 public release

The current authoritative public release checks are the v1.0.1 files. v1.0.0-named dictionary and inventory files are retained as baseline static-browser payload and prior-release provenance artifacts. Previous release DOI details are recorded in reports/release_notes_v1.0.1.md.


Validation and reproducibility

The release includes public validation artifacts so users can inspect the files, counts, and claim-boundary checks supporting the browser.

Artifact Purpose
validation/public_release_checks_v1.0.1.json Current machine-readable v1.0.1 release check summary
validation/public_release_checks_v1.0.1.tsv Current tabular v1.0.1 release check summary
validation/public_language_audit_v1.0.1.tsv Current v1.0.1 public-language, local-link, and DOI-lock audit
reports/table_inventory_v1.0.0.tsv Baseline public table inventory retained in v1.0.1
reports/data_dictionary_v1.0.0.md Baseline field-level data dictionary retained in v1.0.1

The validator also writes validation/checksum_manifest_v1.0.1.tsv during local release checks. Full checksum manifests, larger validation bundles, and archival payloads are preserved through the Zenodo archive.


Public data and maintenance

PrimeReplace Atlas contains public source-derived records and derived transcript-architecture summaries. It does not contain user-submitted data, protected individual-level human subject data, or private clinical records.

The public browser is intended to be maintained at its current URL for at least five years, while immutable release snapshots are archived through Zenodo DOI-versioned records. Future releases should remain versioned through GitHub and Zenodo when the upstream ClinVar/MANE/GENCODE resource stack, data schema, browser, or claim-boundary audit changes.


Main browser pages

  • Overview: index.html
  • Start here: start_here.html
  • Gene search: gene_search.html
  • All-gene index: all_gene_index.html
  • Opportunity units: opportunity_units.html
  • Architecture classes: architecture_classes.html
  • Condition facets: condition_facets.html
  • Methods: methods.html
  • Limitations: limitations.html
  • Citation: citations.html
  • Glossary: glossary.html
  • Downloads: downloads.html

Downloads

The browser provides source tables and validation-linked summaries, including:

  • all_mapped_gene_semantic_summary.tsv
  • all_mapped_low_burden_interpretation_flags.tsv
  • all_mapped_semantic_class_summary.tsv
  • all_mapped_unit_type_summary.tsv
  • allmapped_unique_variant_coordinate_sensitivity.tsv
  • gene_condition_record_groups_v1.tsv
  • gene_condition_record_architecture_recovered_v1.tsv
  • gene_condition_unit_coverage_recovered_v1.tsv
  • condition_architecture_class_summary_recovered_v1.tsv
  • condition_facet_representative_examples_v1.tsv

Large archival payloads, validation manifests, checksums, and release bundles are available through the Zenodo archive.


Data sources

PrimeReplace Atlas uses a frozen public resource stack:

Layer Resource
Genome build GRCh38 / GRCh38.p14
Transcript anchor MANE Select
Transcript structure GENCODE v49
Variant records ClinVar GRCh38 pathogenic/likely pathogenic VCF

ClinVar, MANE, GENCODE, and other third-party public resources remain subject to their original provider terms.


Local browser

The browser is a static site and can also be opened locally.

git clone https://github.com/best916116-crypto/PrimeReplace-Atlas.git
cd PrimeReplace-Atlas
python -m http.server 8000 --directory docs

Then open:

http://localhost:8000/

Repository structure

data/small_tables/
  Compact release tables

docs/
  Static browser pages and browser-downloadable source tables

figures/
  Public release figures and browser-supporting visual assets

reports/
  Release notes, data dictionary, table inventory, and reviewer-facing release documentation

scripts/
  Source scripts used to generate and validate the release

validation/
  Current v1.0.1 release summaries and public-language audit output

Citation

If you use PrimeReplace Atlas in your work, please cite the archived release:

Park J, Cho SI. PrimeReplace Atlas: a transcript-unit architecture database for disease-associated large-fragment genome writing. Zenodo. https://doi.org/10.5281/zenodo.20338345

A manuscript describing the atlas is in preparation. Until the manuscript is available, please cite the Zenodo DOI for the release used in your analysis.


License

This repository uses separate licenses for software and generated data:

  • Code and static browser source: MIT License. See LICENSE.
  • Generated atlas tables, figures, browser-downloadable data files, and archived release materials: Creative Commons Attribution 4.0 International (CC BY 4.0). See LICENSE-DATA.md.

ClinVar, MANE, GENCODE, and other third-party public resources are not relicensed by this repository and remain subject to their original provider terms.


Contact

For public questions, bug reports, or feature requests, please use GitHub Issues.

For release correspondence, contact the release maintainer:

Junjae Park
best9161@korea.ac.kr

About

Transcript-unit architecture database for disease-associated large-fragment genome-writing target review.

Topics

Resources

Stars

2 stars

Watchers

0 watching

Forks

Contributors

Languages