Skip to content

feat: add example ebolavirus mutation pattern configs - #456

Draft
ivan-aksamentov wants to merge 4 commits into
masterfrom
feat/mutation-pattern-analysis
Draft

ivan-aksamentov wants to merge 4 commits into
masterfrom
feat/mutation-pattern-analysis

Conversation

@ivan-aksamentov

@ivan-aksamentov ivan-aksamentov commented May 26, 2026 •

Copy link
Copy Markdown
Member

Placeholder mutationPatterns configuration for all three Orthoebolavirus species to demonstrate the feature

⚠️ event filters, clustering parameters, and motifs are not scientifically validated

Work items

  • Add mutationPatterns config to BDBV (ADAR + APOBEC), EBOV (ADAR + APOBEC CpG), and SUDV (ADAR-only) pathogen.json files with species-specific clustering parameters
  • Add synthetic ADAR-dense and APOBEC-dense example sequences to BDBV
  • Add changelog entries and rebuild unreleased dataset artifacts

@ivan-aksamentov
ivan-aksamentov deployed to refs/pull/456/merge May 26, 2026 10:14 — with GitHub Actions Active
@ivan-aksamentov
ivan-aksamentov marked this pull request as draft May 26, 2026 10:14
Placeholder mutationPatterns configuration for BDBV, EBOV, and SUDV to demonstrate the feature; parameters are not scientifically validated.

- BDBV: ADAR (A>G, T>C) and APOBEC (G>A in YGH context), window 50, cutoff 3, with synthetic example sequences
- EBOV: ADAR (A>G, T>C) and APOBEC-like CpG deamination (C>T in HCG context), window 100, cutoff 5
- SUDV: ADAR (A>G, T>C), window 80, cutoff 4
…ples

- Patterns use `bothStrands` and motifs with the mutated base in parentheses, so each editing signature is one event instead of hand-written complements
- EBOV separates C>T in CpG context from ADAR editing, because they are different mutational processes
- BDBV example sequences start from a tree node sequence, so each one shows one pattern behavior with a result known by construction
…limit

- A C>T or G>A next to a TCW or WGA site must not count as APOBEC3-like, because the substituted base is not the marked base of the site
- Cluster markers count toward the sequence view marker limit, so a sequence with 500 mutations and one cluster shows the coverage-only view
@ivan-aksamentov
ivan-aksamentov force-pushed the feat/mutation-pattern-analysis branch from 5c28551 to 41fd770 Compare September 28, 2026 19:04
@ivan-aksamentov
ivan-aksamentov had a problem deploying to refs/heads/feat/mutation-pattern-analysis September 28, 2026 19:04 — with GitHub Actions Error
@ivan-aksamentov
ivan-aksamentov deployed to refs/pull/456/merge September 28, 2026 19:04 — with GitHub Actions Active
- APOBEC3-like C>T in TCW context drives mpox evolution since the human outbreaks, and these mutations are scattered along the genome, so the pattern reports matches without clusters

This branch was successfully deployed

1 active deployment
refs/pull/456/merge — 21e7cdfa Deployed Sep 29, 2026 by ivan-aksamentov via build-and-deploy-datasets #1545
refs/heads/feat/mutation-pattern-analysis — 21e7cdfa Deployed Sep 29, 2026 by ivan-aksamentov via build-and-deploy-datasets #1544
Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment

Labels

None yet

Projects

None yet

Development

Successfully merging this pull request may close these issues.

1 participant