hi-gene-aso-targets

Evaluates five ASO strategies for up-regulating haploinsufficient genes using splicing, miRNA, and translation data.

1|Updated Oct 25, 2020
One-click install
npx skills add https://github.com/bfairkun/dotfiles --skill hi-gene-aso-targets-bfairkun
Or copy as Structured Prompt for Agent▼
Please help me install this Agent Skill.
Skill: hi-gene-aso-targets
Source: https://github.com/bfairkun/dotfiles/tree/main/agents/.agents/skills/hi-gene-aso-targets
Command: npx skills add https://github.com/bfairkun/dotfiles --skill hi-gene-aso-targets-bfairkun

SYSTEM DOCUMENTATION & REQUIREMENTS

What problem does it solve? Finding antisense oligonucleotide (ASO) targets that raise the dosage of a haploinsufficient gene requires evaluating many orthogonal mechanisms, and most candidates turn out negative. This Skill provides a systematic, judgement-driven framework for reaching a defensible answer across five strategies rather than chasing a single positive hit. ## Core Features & Use Cases - Unproductive splicing analysis: Uses SplisER SSE (not leafcutter PSI) to quantify NMD-coupled isoforms and decide whether redirecting splicing is worthwhile based on basal splice-site usage. - 3'UTR miRNA site screening: Combines conserved and non-conserved TargetScan tables, gates sites by tissue miRNA expression, and ranks candidates on context++ scores. - RiboNN translation walk: Tiles 20-nt deletions across UTRs to find repressive elements, interpreting ΔTE against a MANE reference distribution. - Small-molecule panel: Queries a dose-response database for expression changes, with explicit tests to rule out intron-retention artifacts from splicing inhibitors. - Use Case: Given a gene like ATP6V0C, produce a fully documented notebook covering all five strategies, including negative results, cross-referenced against phyloP conservation and visualized in an IGV.js browser. ## Quick Start Ask the assistant to find ASO targets for raising the dosage of a specific gene, providing the gene symbol and the relevant cell type or tissue.

Frequently Asked Questions about hi-gene-aso-targets

High-intent search queries and answers about installing and using this skill.

FAQPage Schema
How do I find ASO targets to increase expression of a haploinsufficient gene?▼

Evaluate five orthogonal strategies: unproductive splicing redirection, 3'UTR miRNA site blocking, uORF and 5'UTR element targeting, RiboNN-predicted repressive elements, and small-molecule up-regulation. Most genes come out negative, so document negative results fully alongside any candidates.

Should I use leafcutter PSI or SplisER SSE to quantify splice site usage?▼

Use SplisER SSE for measuring how much unproductive splicing exists to redirect. PSI is a within-cluster fraction that can be contaminated by neighbouring genes' introns, making apparent changes track expression ratios rather than true splicing changes.

How do I rank TargetScan miRNA sites for ASO targeting?▼

Gate sites by miRNA expression in the relevant tissue first, then rank survivors on raw context++ scores rather than percentiles, which are not comparable across miRNAs. Use both the conserved and non-conserved TargetScan tables, since short 3'UTRs are mostly non-conserved sites.

Why does featureCounts undercount genes with overlapping annotations?▼

featureCounts without the -O flag discards reads overlapping multiple annotated features as Unassigned_Ambiguity rather than reassigning them. Sanity-check gene-level counts against junction counts before trusting any CPM values for overlapping loci.

What are the limitations of RiboNN for finding repressive UTR elements?▼

RiboNN does not support N-masking, so 20-nt deletions must be tiled across UTRs instead, and runs are slow. Interpret ΔTE against a reference distribution rather than zero, since a gene already at maximum baseline translation efficiency has no headroom.

When is an NMD-inducible splice site not worth targeting with an ASO?▼

A site is not worth targeting when its basal splice-site usage is below roughly 5-10% of the canonical site, even if it shows a strong NMD induction signature. A large fold-change from a near-zero baseline yields too little absolute product to matter therapeutically.