Replication and colocalization refine gene prioritization in vascular dementia
Alzheimer's & Dementia: Translational Research & Clinical Interventions · 2026-10 · 12 (4)
DOI: 10.1002/trc2.70348 ↗Kurz gefasst
Fachartikel in der Zeitschrift Alzheimer's & Dementia: Translational Research & Clinical Interventions, online erschienen am 18.11.2026. Verfasst von Yumin Zhou und 5 weiteren Autor:innen. Die Arbeit nennt 22 Literaturangaben. Der Volltext ist frei zugänglich (Open Access, CC BY-NC-ND 4.0).
Themen: Demenz
Abstract (Originalsprache)
INTRODUCTION Molecular quantitative trait locus analysis can nominate genes for vascular dementia (VaD), but association alone does not identify the transcript underlying a disease signal. We assessed candidates using independent outcome replication and regional analyses. METHODS We used summary‐data‐based Mendelian randomization (SMR) with eQTLGen Phase I blood expression data and FinnGen R13 VaD. Of 19,250 annotated probes, 15,785 were eligible before outcome matching and 15,631 formed the tested family after outcome quality control. Experiment‐wide discoveries underwent independent outcome replication in UK Biobank (GCST90473240), regional colocalization, heterogeneity in dependent instruments (HEIDI) testing, and BrainMeta version 2 cortex follow‐up. RESULTS Kinesin light chain 3 (KLC3) was the only experiment‐wide significant signal ( p = 9.79×10 −8 ); LON peptidase N‐terminal domain and ring finger 1 (LONRF1) met false discovery rate correction only. KLC3 did not meet the UK Biobank replication criterion (SMR effect estimate = 0.095, 95% CI: −0.214 to 0.404; p = 0.547). The KLC3 expression quantitative trait locus peak (rs9749618) differed from the FinnGen VaD peak (rs429358, apolipoprotein E [APOE] ε4). Under a model allowing one causal variant per trait, colocalization favored distinct regional signals (PP3≈1; negligible PP4 across tested priors). Blood HEIDI results differed between implementations; cortex analyses did not provide convergent transcript support. LONRF1 colocalization was sensitive to the prior. No candidate met all criteria in the study‐specific evidence framework. DISCUSSION Independent outcome replication and regional analyses refined interpretation of the expression screen. Under the current data and models, KLC3 was not supported as the transcript underlying the VaD signal. Non‐replication does not establish absence of KLC3 biology, and the complex APOE region requires cautious attribution.
Zitieren
APA
Zhou, Y., Gong, S., Li, Q., Lv, D., Yao, J., & Bi, X. (2026). Replication and colocalization refine gene prioritization in vascular dementia. Alzheimer's & Dementia: Translational Research & Clinical Interventions, 12(4). https://doi.org/10.1002/trc2.70348
Vancouver
Zhou Y, Gong S, Li Q, Lv D, Yao J, Bi X. Replication and colocalization refine gene prioritization in vascular dementia. Alzheimer's & Dementia: Translational Research & Clinical Interventions. 2026;12(4). doi:10.1002/trc2.70348
BibTeX
@article{Zhou2026,
author = {Zhou, Yumin and Gong, SiHao and Li, Qian and Lv, Dandan and Yao, JinFeng and Bi, Xia},
title = {Replication and colocalization refine gene prioritization in vascular dementia},
journal = {Alzheimer's & Dementia: Translational Research & Clinical Interventions},
year = {2026},
volume = {12},
number = {4},
doi = {10.1002/trc2.70348}
}
EUVIMED