rewire.it
model · family

Pangolin

Pangolin predicts changes in splice-site strength from DNA variants. It accepts variant files or custom sequence inputs.

0 evaluations · 0 metric rows

Shared profile: Pangolin. This page retains the exact record and its evaluation context.

At a glance

Explanatory profile: source reviewed · Automated source review, 2026-09-16. This does not change the review status of its results.

Inputs, outputs and configuration
PropertyDescription and evidence
MaskingDefault mask=True; a mask=False evaluation is a distinct configurationtkzeng/Pangolin official source · README.md: introduction and Usage (command-line), including supported variants and --mask
Configuration in this recordpublished checkpointstkzeng/Pangolin official source · README.md: introduction and Usage (command-line), including supported variants and --mask
Model typeNot extracted or verified for this record.
Training dataNot extracted or verified for this record.
Context limitsNot extracted or verified for this record.
AccessNot extracted or verified for this record.
Code licenceNot extracted or verified for this record.
Weights licenceNot extracted or verified for this record.

How it works

Conceptual procedure

Schematic of the documented input, computation and output; not an executable configuration.

Conceptual procedureDNA context. Then: Dilated residual convolutions. Then: Tissue-specific outputs. Then: Splice strength. Then: Variant-induced changeDNA contextDilated residual convolutionsTissue-specific outputsSplice strengthVariant-induced change
Read the diagram as text
  1. DNA context
  2. Dilated residual convolutions
  3. Tissue-specific outputs
  4. Splice strength
  5. Variant-induced change
tkzeng/Pangolin official source · Original paper linked in README: https://doi.org/10.1186/s13059-022-02664-4, Figure 1 and Methods: Deep neural network architecture; README Usage

Architecture

Pangolin uses 16 residual blocks with dilated convolutions and skip connections. Separate outputs estimate splice-site probability and usage across heart, liver, brain and testis. The published model was trained using sequence and splicing measurements from human, rhesus macaque, rat and mouse.

tkzeng/Pangolin official source · Original paper linked in README: https://doi.org/10.1186/s13059-022-02664-4, Methods: Deep neural network architecture; Results: Pangolin predicts splice site usage

Reference genome and transcript annotation define the sequence context. The neural predictor estimates splice-site strength; the command-line tool reports the largest positive and negative changes near each variant. Masking optionally removes particular gains and losses at annotated sites.

tkzeng/Pangolin official source · README.md: introduction and Usage (command-line), including supported variants and --mask

Benchmarks and results

Release 2026-09-16-d74d282221a9 · 0 evaluations · 0 metric rows. Different protocols are not a single leaderboard.

No evaluations linked in this release.

Strengths and limitations

Strengths supported by sources

  • Provides changes in splice strength and their positions, with configurable scoring distance and masking.tkzeng/Pangolin official source · README.md: introduction and Usage (command-line), including supported variants and --mask

Limitations and conditions

  • Only substitutions and simple indels are supported by the documented interface. Missing gene annotations, reference mismatches and chromosome-edge cases can exclude variants.tkzeng/Pangolin official source · README.md: introduction and Usage (command-line), including supported variants and --mask
Profile review details

Primary project documentation or paper inspected for the explanatory claims and cited locations. Reviewed coverage concerns this narrative, not complete metadata, independent reproduction or a performance ranking.

Stable record: catalog-model-pangolin

Sources and history

Release 2026-09-16-d74d282221a9 · Record review: discovered

Download this release
Technical metadata and extraction receipts

Stable ID: catalog-model-pangolin

areas
dna-genomes
method types
specialist
entity level
family
version
published checkpoints
reported name
Pangolin
access
Public specialist code and models under GPL-3.0.
method type
specialist
missing metadata
checkpoint revision: not_yet_extracted; training data: not_yet_extracted; licence: not_yet_extracted
Related records

Suggest a correction