TripletDGC: Assessing critical cell types of disease genes by integrating single-cell genomics and human genetics

Advances in human genetics have enabled the investigation of the genetic architecture underlying numerous complex traits. Many publicly available databases establish a binary relationship between genes and diseases. However, the limitations of this binary framework hinder a deeper understanding of how disease-related genes influence cell types and cell states during disease onset.

To solve the problems, a research team led by Bingbo Wang published their new research on 15 October 2025 in Frontiers of Computer Science co published by Higher Education Press and Springer Nature.

The team constructed a ternary dataset by integrating single-cell sequencing data with human genetic data: TripletDGC. This dataset establishes a triplet relationship between disease genes, diseases, and cell types, highlighting the critical cell types associated with disease genes and offering deeper insights into the underlying disease mechanisms.

In this study, to identify the critical cell types associated with disease genes, the team integrated single-cell gene expression profiling with eQTL regulation data. They first mapped SNPs to gene regions to identify genes regulated by disease-related genes. These genes were then enriched and analyzed using marker genes from the cell types identified in the single-cell expression data, ultimately revealing the critical cell types linked to disease genes. This comprehensive analysis allowed the team to not only pinpoint the genes regulated by disease-related variants but also uncover the specific cell types most strongly linked to disease mechanisms, providing deeper insights into the molecular basis of diseases.

The statistical analysis revealed an uneven distribution of disease genes across diseases, with most diseases linked to only a few genes, highlighting disease specificity. Less than 10% of disease-gene associations showed statistically significant links to critical cell types, likely due to gene pleiotropy (average of 2.23 diseases per gene) and incomplete regulatory mechanisms. On average, a single disease gene is connected to just 1.37 critical cell types, while a disease is associated with 11.7 cell types through 55.90 different genes. This underscores the importance of analyzing individual disease genes to accurately capture cell state features. Additionally, each cell type is linked to an average of 8.54 diseases, suggesting shared state characteristics that may explain potential comorbidities.

TripletDGC delineates the association between disease genes and diseases by characterizing specific cell types, offering novel insights into disease mechanisms within the feature space defined by single-cell data. This resource provides a strong foundation for drug target prediction and the discovery of new disease genes.