GWAS of plasma pTau181 in East Asian cohort

PRIMARY OUTCOME

plasma pTau181 levels

EXPECTED OUTCOMES

1. **APOE ε4 effect on pTau181**: APOE ε4 carriers will show plasma pTau181 levels 1.5-2.0 pg/mL higher than non-carriers (β = 0.38, SE = 0.05, p = 2.3×10⁻¹²), consistent with ~25-35% increase relative to mean. 2. **KCNJ3 region association**: At least one SNP in the KCNJ3 locus will reach suggestive significance (p < 1×10⁻⁶), with expected effect size of β = 0.15-0.25 SD per allele (approximately 0.3-0.5 pg/mL per copy of effect allele). 3. **Genome-wide significance hits**: 3-7 SNPs will achieve genome-wide significance (p < 5×10⁻⁸), including known AD loci (APOE, BIN1, CLU) that influence pTau181 as an endophenotype. 4. **Population-specific signals**: 1-3 novel loci will show genome-wide significance specifically in this East Asian cohort, with allele frequencies differing from European populations by ≥10%. 5. **Heritability estimate**: SNP-based heritability (h²SNP) for plasma pTau181 will be estimated at 0.12-0.22 (SE = 0.04), explaining 12-22% of phenotypic variance. 6. **QPCTL and other pTau181-linked loci**: Loci previously associated with pTau181 (QPCTL, GBA) will replicate with effect sizes of β = 0.20-0.35 SD per allele in this cohort. 7. **Correlation with CSF biomarkers**: Plasma pTau181 will correlate with CSF pTau181 (Spearman ρ = 0.55-0.70, p < 1×10⁻¹⁰) and with CSF total tau (ρ = 0.45-0.60) in the subset with paired CSF samples. ---

SUCCESS CRITERIA

- **GWAS significance threshold**: Primary analysis achieves at least 2 loci meeting genome-wide significance (p < 5×10⁻⁸) after multiple testing correction, with λGC < 1.05 indicating adequate population stratification control. - **Effect size validation**: Identified GWAS signals demonstrate effect sizes with 95% CI excluding the null, and β estimates are within 1.5-fold of previously reported values for the same variants. - **Replication requirement**: Top 3 SNPs replicate in independent East Asian cohort (N ≥ 500) with p < 0.05 and consistent direction of effect, demonstrating external validity. - **APOE genotype accuracy**: APOE genotyping shows 100% concordance between TaqMan assay and direct sequencing in a random 5% subset (N = 150), with Hardy-Weinberg equilibrium maintained (p > 0.01). - **pTau181 assay performance**: Inter-plate CV <15%, intra-plate CV <10%, and spike-recovery 85-115% across the standard curve range (5-200 pg/mL). - **Sample quality**: DNA samples show A260/280 ≥1.8, A260/230 ≥1.9, and median call rate ≥99% on genotyping arrays after QC filtering. - **PRS predictive value**: Polygenic risk score constructed from top GWAS SNPs explains ≥5% of variance in plasma pTau181 (R² ≥ 0.05, p < 1×10⁻⁶) in held-out test set (N = 500).

PROTOCOL

--- ### PHASE 1: SAMPLE COLLECTION AND DNA EXTRACTION (Week 1-4) **1.1 Study Population** - Cohort: K-ROAD (Korean ROAD to Alzheimer's Disease) - Sample size: N = 3,000 East Asian participants - Inclusion: Age ≥60 years, willing to provide informed consent, East Asian ancestry (self-reported) - Exclusion: Active malignancy, end-stage renal disease, recent major surgery **1.2 Sample Collection** - Fasting blood draw (10 mL EDTA whole blood, 10 mL plasma) - Centrifuge at 1,500 × g for 15 minutes at 4°C within 1 hour of collection - Aliquot plasma (500 μL × 4 tubes) and store at -80°C - Store buffy coat at -80°C for DNA extraction **1.3 DNA Extraction** - Method: Magnetic bead-based extraction (Qiagen QIAamp DNA Blood Mini Kit or equivalent) - Equipment: QIACube Connect (Qiagen) or equivalent automated system - Reagents: - Qiagen Protease (20 mg/mL) - Buffer AL ( guanidinium hydrochloride-based) - Buffer AW1 and AW2 (ethanol-containing wash buffers) - Buffer AE (10 mM Tris-Cl, 0.5 mM EDTA, pH 9.0) - DNA concentration: Nanodrop 2000 (260/280 ratio ≥1.8, 260/230 ≥1.9) - Quantification: Qubit 3.0 Fluorometer (dsDNA BR Assay Kit) - Store DNA at 4°C (working) and -80°C (long-term) --- ### PHASE 2: GWAS GENOTYPING (Week 5-12) **2.1 Genotyping Platform** - Array: Illumina Asian筛查 Array (ASA) or Affymetrix Axiom Precision Medicine Research Array (PMRA) - Target: ~700,000 SNPs with imputation to ~10 million SNPs - Coverage: Focus on KCNJ3 region (chr4: 81,500,000-83,500,000 GRCh37) and APOE region (chr19: 44,900,000-45,100,000) **2.2 APOE Genotyping (Direct)** - Method: TaqMan SNP Genotyping Assay - SNPs: rs7412 (APOE ε2), rs429358 (APOE ε4) - Reagents: - TaqMan GTXpress Master Mix (Applied Biosystems) - APOE rs7412 Assay: C____904973_10 - APOE rs429358 Assay: C____3084793_20 - Equipment: QuantStudio 7 Flex Real-Time PCR System - Thermal cycling: 95°C 20s → (95°C 3s, 60°C 30s) × 40 cycles **2.3 Quality Control** - Sample-level QC: - Call rate <98%: exclude - Gender discrepancy: verify with XY markers - Cryptic relatedness (π > 0.2): remove one sample per pair - Heterozygosity outliers (>3 SD from mean): flag for review - SNP-level QC: - Call rate <95%: exclude - HWE p < 1×10⁻⁶: exclude (except APOE region documented) - MAF <1%: exclude for common GWAS; include for rare variant analysis **2.4 Imputation** - Reference panel: 1000 Genomes Phase 3 (GRCh37) or TOPMed r2 - Software: IMPUTE2 or Michigan Imputation Server - Pre-phasing: SHAPEIT2 - Post-imputation filter: INFO score >0.4 --- ### PHASE 3: PLASMA pTAU181 QUANTIFICATION (Week 13-16) **3.1 Assay Platform** - Method: Simoa HD-X Analyzer (Quanterix) with Homebrew assay - Alternative: Fujirebio Lumipulse G pTau181 (automated CLEIA) - Target biomarker: Phosphorylated tau at threonine 181 **3.2 Simoa pTau181 Assay (Homebrew)** - Capture antibody: Anti-pTau181 (Rabbit monoclonal, clone: E7A1A, Cell Signaling Technology) - Working concentration: 3 μg/mL in PBS - Coating: 50 μL/well in 96-well Quanterix plates - Incubation: overnight at 4°C, then block with 0.1% BSA/PBS - Detection antibody: Biotinylated anti-total tau (DAKO) - Working concentration: 0.3 μg/mL in assay diluent - Incubation: 30 minutes at 30°C with shaking - Enzyme: Streptavidin-β-galactosidase (SBG) - Dilution: 1:100 in SBG diluent - Incubation: 5 minutes at 30°C - Substrate: Resorufin β-D-galactopyranoside (RGP) - Concentration: 100 μM in substrate buffer - Incubation: 60 seconds at 30°C - Stop: 0.1 M EDTA, pH 8.0 **3.3 Internal Quality Controls** - Low QC: Pooled plasma from AD patients (target: 15-20 pg/mL) - Medium QC: Pooled plasma (target: 40-50 pg/mL) - High QC: Spiked plasma (target: 80-100 pg/mL) - Acceptance criteria: CV <20% for all QC levels - Run acceptance: ≥80% of QC samples within 2 SD of target **3.4 Sample Analysis** - Run duplicate measurements per sample - Minimum sample volume: 100 μL plasma per replicate - Dilution: 1:4 in sample diluent (if needed for high samples) - Background subtraction: Average of blank wells (no sample) --- ### PHASE 4: STATISTICAL ANALYSIS (Week 17-24) **4.1 Primary Analysis** - Software: PLINK 2.0, R 4.2+ (data manipulation), SNPTest (imputed data) - Model: Linear regression with plasma pTau181 as continuous dependent variable - Covariates: - Age (continuous) - Sex (categorical) - APOE ε4 status (0, 1, or 2 copies) - PC1-PC5 (principal components for ancestry) - Batch (genotyping plate) - Additive genetic model assumed **4.2 Population Stratification** - PCA: SmartPCA (EIGENSOFT 7.2) - Compute 10 ancestry PCs using LD-pruned SNPs (r² < 0.1) - Visual inspection: Q-Q plots and Manhattan plots - λ (genomic inflation factor) target: 1.0-1.05 **4.3 Region of Interest Analysis** - KCNJ3: ±100 kb around gene body - cis-pQTL analysis: Test all SNPs within 1 Mb of KCNJ3 transcription start site - Conditional analysis: Adjust for top GWAS hit to identify secondary signals **4.4 APOE-Stratified Analysis** - Stratify by APOE ε4 status: - Group 1: ε4 non-carriers (ε2/ε2, ε2/ε3, ε3/ε3) - Group 2: ε4 heterozygotes (ε3/ε4, ε2/ε4) - Group 3: ε4 homozygotes (ε4/ε4) - Interaction analysis: SNP × APOE ε4 status **4.5 Power Calculation** - Assumptions: - N = 3,000 - α = 5×10⁻⁸ (genome-wide) - MAF ≥0.05 - Effect size (β): 0.1-0.2 SD per allele - Target power: ≥80% --- ### PHASE 5: FUNCTIONAL VALIDATION IN SILICO (Week 25-28) **5.1 eQTL Analysis** - Datasets: Japanese MetaboBank, K-ROAD RNA-seq (if available) - Tools: Matrix eQTL, QTLtools - Test: Association between KCNJ3 SNPs and KCNJ3 mRNA expression in blood/brain **5.2 Bioinformatics Annotation** - Variant effect prediction: SIFT, PolyPhen-2, CADD - Regulatory annotation: ENCODE, Roadmap Epigenomics - Brain expression: GTEx v8 (brain tissue), Allen Brain Atlas - Pathway analysis: MAGMA, DEPICT **5.3 Mendelian Randomization** - Software: TwoSampleMR (R package) - Instrumental variables: Independent GWAS-significant SNPs (r² < 0.01) - Outcome: Alzheimer's disease (IGAP summary statistics) - Sensitivity: Heterogeneity, pleiotropy tests --- ### PHASE 6: REPLICATION AND REPORTING (Week 29-36) **6.1 Replication Cohort** - Internal: K-ROAD independent subset (N = 1,000) - External: East Asian cohort (e.g., Japanese-ADNI, Shanghai Aging Study) - Replication threshold: p < 0.05 with consistent direction of effect **6.2 Multi-marker Analysis** - Polygenic risk score (PRS): PRSice-2 - Base: East Asian AD GWAS (if available) or European AD GWAS with LD reference - Target: Plasma pTau181 levels - Clumping: r² = 0.1, kb = 250 **6.3 Data Sharing** - Summary statistics upload: GWAS Catalog, SSGAC - Individual-level data: Controlled access repository (dbGaP or equivalent) ---

LINKED HYPOTHESES

HTML iframe

<iframe src="http://scidex.ai/artifact/exp-792473c3-ae88-48b9-9be1-bb38969946ee?embed=1" width="100%" height="600" style="border:0;border-radius:8px"></iframe>

Markdown link

[GWAS of plasma pTau181 in East Asian cohort](http://scidex.ai/artifact/exp-792473c3-ae88-48b9-9be1-bb38969946ee)

Direct URL

http://scidex.ai/artifact/exp-792473c3-ae88-48b9-9be1-bb38969946ee

Preview