Sullivan P et al., The American Journal of Human Genetics - A concise walkthrough of data-driven heuristics and a splicing checklist derived from large-scale exon, branchpoint, and experimentally validated variant analyses to improve interpretation of splice-altering variants. Key terms: splicing, splice-altering variants, heuristics, SpliceVarDB, pseudoexon.

Study Highlights:
The authors analyzed ~202,000 canonical protein-coding exons and 19,034 experimentally validated branchpoints to define sequence, spacing, and motif-strength requirements for U2 splicing, finding 95.9% of exons met these criteria. Using 11,860 experimentally validated single-nucleotide variants from SpliceVarDB, they derived data-driven heuristics, subgroups, and a spliceogenicity metric to predict whether variants alter splicing. Heuristics were implemented as donor (DD) and acceptor (DA) flowcharts plus an in silico checklist, applied to pseudoexon formation, and used to quantify typical splicing outcomes. The study notes limitations including sparse examples for splicing regulatory elements and context- or tissue-specific effects not fully captured.

Conclusion:
Empirically grounded heuristics and a practical checklist bridge biological splicing rules and in silico prediction, improving evaluation of putative splice-altering variants while acknowledging limits from incomplete SRE data and context dependence; ongoing contribution to SpliceVarDB and integration into tools should refine performance.

QC:
This episode was checked against the original article PDF and publication metadata for the episode release published on 2025-04-16.

QC Scope:
- article metadata and core scientific claims from the narration
- excludes analogies, intro/outro, and music

QC Summary:
- factual score: 10/10
- metadata score: 10/10
- supported core claims: 7
- claims flagged for review: 0
- metadata checks passed: 4
- metadata issues found: 0

Metadata Audited:
- article_doi
- article_title
- article_journal
- license

Factual Items Audited:
- Data-driven splicing heuristics framework replaces black-box AI for SAV interpretation and uses a transparent checklist grounded in splicing biology.
- Minimal splicing requirements were defined from extensive analysis, with 95.9% of exons meeting these criteria.
- SpliceVarDB provided data for over 11k experimentally validated SAVs, used to derive heuristics with a minimum threshold of 10 variants per heuristic.
- Disruption of donor and acceptor sites is organized into DD and DA flowcharts to classify spliceogenicity, with subgroups.
- Splicing outcomes are quantitatively distributed: exon skipping ~51.3%, exon truncation ~36.3%, intron retention ~17.4%, exon extension ~16.2%.
- Deep intronic variants can create pseudoexons; the checklist flags 82% of confirmed pseudoexons.

QC result: Pass.