# GSE272233 DMD CRISPR-correction transcriptome reference

## Scope

This asset summarizes bulk RNA-seq from 21 samples: wild-type myotubes plus three DMD exon-duplication backgrounds and their corresponding CRISPR-corrected clones, each with three replicates.

The primary contrasts are corrected versus untreated within each mutation
background. Untreated-versus-WT contrasts are retained only for an
exploratory reversal check because the WT line is not isogenic.

## Processing

- Input SHA-256: `072e1398918140af4898cc895dcbd6c72c28beb39353834063e457f9d0161eee`.
- Raw gene rows: 62757.
- Aggregated symbols/identifiers: 62710.
- Genes retained by `edgeR::filterByExpr`: 22230.
- Normalization: edgeR TMM.
- Inference: edgeR quasi-likelihood negative-binomial GLM with robust dispersion.
- Multiple testing: Benjamini-Hochberg FDR within each contrast.

## Correction contrasts

| contrast | tested genes | FDR < 0.05 | FDR < 0.05 and |log2FC| >= 1 |
| --- | ---: | ---: | ---: |
| dup2_correction | 22230 | 2812 | 1090 |
| dup2_9_correction | 22230 | 375 | 167 |
| dup8_9_correction | 22230 | 2736 | 979 |

## Candidate interpretation

- Current Perturbation Lab candidates observed in the filtered matrix: 21/21.
- Candidates with at least one correction contrast at FDR < 0.05: 5.
- Candidates with an exploratory disease-to-correction reversal in at least one context: 2.

A candidate-level value describes how that gene's transcript changes after
editing the DMD locus. It is not the effect of perturbing that candidate.

## Claim boundary

GSE272233 is used as a bulk transcriptomic correction reference and GSE293514 as a bounded external myoblast context screen. Neither dataset supplies direct, independent DMD perturbation validation for the current 21 candidate CRISPRi perturbations, observed double perturbations, CRISPRa/overexpression responses or perturbed fate trajectories.