T2D NK Cell Composition in PBMCs
NK cells are part of the innate lymphoid compartment in PBMCs. Their abundance in type 2 diabetes PBMC samples is inconsistently reported across studies.
Tkachenko et al. 2024 Finding
Tkachenko et al. 2024 (medRxiv preprint) performed DCATS differential composition analysis on PBMC scRNA-seq from 2 T2D and 2 healthy controls (Russian cohort, all female). NK cells (cluster 3) were significantly less prevalent in T2D samples compared to controls. The authors explicitly note this contrasts with Gu et al. 2024. ^[extracted] (from Tkachenko et al. 2024)
Gu et al. 2024 Finding
Gu et al. 2024 (Korean cohort, 293,923 PBMCs) reported higher NK cell proportions in T2D relative to non-diabetes controls, alongside higher CD16 monocyte proportions. ^[extracted] (from Gu et al. 2024)
Possible Explanations for Discrepancy
- Sample size: Tkachenko et al. scRNA-seq has n=2 per group, limiting power. The authors themselves suggest this may account for the divergent NK finding. ^[extracted] (from Tkachenko et al. 2024)
- Cohort differences: Russian vs. Korean populations may differ in baseline NK cell abundance or T2D-associated NK dynamics.
- Sex composition: Tkachenko scRNA-seq subset was all female; Gu et al. cohort sex distribution is not fully reported.
- Technical factors: Different scRNA-seq platforms, chemistry versions, clustering resolution, and annotation references may produce different cell-type assignments.
Relevance to Project
- The NK cell discrepancy between these two PBMC scRNA-seq studies illustrates the cross-study heterogeneity challenge even within the same assay type (PBMC scRNA-seq).
- For the PBMC ancestry project, NK cell composition should be treated as a variable that may differ by cohort, ancestry, or disease status — and any ancestry-associated NK signal should be validated against study-level confounders.