What question did this study set out to answer?

The aim is to design and validate specific regulatory elements for Müller glia using deep learning techniques.

July 12, 20260 citationsOpen Access

Deep learning design and in vivo validation of Müller glia-specific cis-regulatory elements: code and data

OFOriol Fornés KKKyung Duk Koh RMRebecca M. Marton

Key Points

The aim is to design and validate specific regulatory elements for Müller glia using deep learning techniques.
Generated processed mouse retinal multiome (RNA + ATAC) matrices with cell-type annotations.
Developed a computational pipeline using Seurat, ArchR, and gReLU for data processing and element design.
Outputs include AnnData objects, selected sequences, and evolved elements.
Produced core AnnData (.h5ad) objects and various selected sequences for Müller glia.
Identified non-redundant transcription factor motif clusters from the retinal multiome data.

Abstract

Code and data used to generate the results for the publication: "Deep learning design and in vivo validation of Müller glia-specific cis-regulatory elements". Summary data/: Processed mouse retinal multiome (RNA + ATAC) matrices, cell-type annotations, and non-redundant transcription factor motif clusters. outputs/: Core AnnData (.h5ad) objects, selected sequences, and evolved elements. src/: Step-by-step computational pipeline notebooks (from data processing via Seurat/ArchR to gReLU model training, element design, and figure generation). The computational environment is managed via Pixi (src/pixi.toml).

问 AI

Bookmark

View Full Paper