Qadir, M. M. F. | Search bioRxiv

bioRxiv2025

Supervised machine learning identifies impaired mitochondrial quality control in β cells with development of type 2 diabetes

In type 2 diabetes (T2D), molecular pathways driving {beta} cell failure are difficult to resolve with standard single cell analysis. Here we developed an interpretable, supervised machine learning framework that couples sparse rule-based classification (SnakeClassifier), pathway constrained modelling (BlackSwanClassifier), and {beta} cell mitochondrial fitness stratification (Kolmogorov-Arnold Neural Networks KANN), linking and integrating them into disease mechanisms in single cell RNA sequencing (scRNA-seq) from 52 human donors. SnakeClassifier trained on 50 genes accurately predicted T2D at single cell resolution, outperforming classical ensemble machine learning classifier models, and yielded donor level diabetes scores that correlated with chronic hyperglycemia. The clustering of {beta} cell populations ({beta}1-4) revealed a resilient non-diabetic (ND) {beta}1 subtype characterized by preserved {beta} cell identity genes and lower disease risk, whereas T2D {beta}2-4 subtypes exhibited upregulation of genes involved in cellular and mitochondrial stress and suppression of genes promoting oxidative phosphorylation and insulin secretion. Mitophagy emerged as the dominant program linked to T2D and a mitophagy focused BlackSwanClassifier nominated PINK1, BNIP3, and FUNDC1 as key regulators. PINK1 was enriched in ND {beta}1, decreased with T2D disease score and connected sex stratified mitophagy. We generated a KANN derived mitochondrial fitness index (MFI) integrating mitophagy, mitochondrial proteostasis, biogenesis and oxidative phosphorylation into a single interpretable score (R2 = 0.934 vs module-based mitochondria quality index), which identified mitophagy PINK1, SQSTM1, PRKN and BNIP3 as top contributors to T2D progression. These transparent models unify prediction with T2D disease mechanism and identify the mitophagy receptor PINK1 as a central determinant of {beta} cell metabolic fitness

systems biology↗

bioRxiv2024

Single cell regulatory architecture of human pancreatic islets suggests sex differences in β cell function and the pathogenesis of type 2 diabetes.

Biological sex affects the pathogenesis of type 2 and type 1 diabetes (T2D, T1D) including the development of {beta} cell failure observed more often in males. The mechanisms that drive sex differences in {beta} cell failure is unknown. Studying sex differences in islet regulation and function represent a unique avenue to understand the sex-specific heterogeneity in {beta} cell failure in diabetes. Here, we examined sex and race differences in human pancreatic islets from up to 52 donors with and without T2D (including 37 donors from the Human Pancreas Analysis Program [HPAP] dataset) using an orthogonal series of experiments including single cell RNA-seq (scRNA-seq), single nucleus assay for transposase-accessible chromatin sequencing (snATAC-seq), dynamic hormone secretion, and bioenergetics. In cultured islets from nondiabetic (ND) donors, in the absence of the in vivo hormonal environment, sex differences in islet cell type gene accessibility and expression predominantly involved sex chromosomes. Of particular interest were sex differences in the X-linked KDM6A and Y-linked KDM5D chromatin remodelers in female and male islet cells respectively. Islets from T2D donors exhibited similar sex differences in differentially expressed genes (DEGs) from sex chromosomes. However, in contrast to islets from ND donors, islets from T2D donors exhibited major sex differences in DEGs from autosomes. Comparing {beta} cells from T2D and ND donors revealed that females had more DEGs from autosomes compared to male {beta} cells. Gene set enrichment analysis of female {beta} cell DEGs showed a suppression of oxidative phosphorylation and electron transport chain pathways, while male {beta} cell had suppressed insulin secretion pathways. Thus, although sex-specific differences in gene accessibility and expression of cultured ND human islets predominantly affect sex chromosome genes, major differences in autosomal gene expression between sexes appear during the transition to T2D and which highlight mitochondrial failure in female {beta} cells.

genomics↗