Diabetic Retinopathy Candidate Genes from Transcriptomic and Single-Cell Analysis

A table of candidate genes associated with de novo lipogenesis and diabetic retinopathy, identified through integrated bioinformatics. The dataset includes results from transcriptomic analysis, single-cell RNA sequencing, and GraphBAN modeling for gene-compound interactions. It was published by Zhenghong Jin on figshare in April 2026.

Use Cases

• Validate the identified candidate genes AHR and SLC1A5 in independent diabetic retinopathy cohorts.
• Analyze the functional enrichment results for pathways implicated in diabetic retinopathy pathogenesis.
• Use the GraphBAN model predictions to screen for potential compounds interacting with the AHR and SLC1A5 genes.
• Map the single-cell RNA sequencing expression patterns of candidate genes to specific cell populations like cDC2 and classical monocytes.
• Assess the predictive performance of the nomogram model for diabetic retinopathy risk based on gene expression scores.

Strengths

• Integrates multiple data types: public transcriptomics, single-cell RNA-seq, and computational GraphBAN modeling.
• Identifies two specific candidate genes (AHR and SLC1A5) with RT-qPCR validation in patient samples.
• Provides functional enrichment analysis linking genes to specific biological pathways.

Limitations

• Dataset is very small at 26.5 KB, containing summary results rather than raw sequencing data.
• Sample size for validation (DR patients vs. controls) is not specified, limiting statistical assessment.
• The single-cell analysis is focused on DR fibrous membranes, which may not represent all retinal tissues.

Provenance

Source

Public databases for DR-related transcripts and DNL-associated genes, with validation via RT-qPCR.

Collection Method

Integrated bioinformatics analysis, single-cell RNA sequencing, and computational GraphBAN modeling.

Freshness

Last updated in April 2026.