rs429358

A case-control study of 335 generalized aggressive periodontitis (GAgP) patients and 114 healthy controls identified 4 SNPs significantly associated with GAgP risk: MMP8 rs11225395 (OR=1.40), EGF rs2237051 (OR=1.41), PPARα rs4253623 (OR=1.53), and APOE rs429358 (OR=1.79). A multi-locus genetic risk score based on these variants improved disease prediction, with each additional risk allele conferring ~50% increased risk (OR=1.50), and the GRS combined with conventional risk factors improved discrimination from AUC 0.695 to 0.740.

This study examined 4960 SNP triples from five genes in the APOE region (BCAM, NECTIN2, TOMM40, APOE, APOC1) in 2789 Alzheimer's disease cases and 16,334 controls using a novel co-skewness metric to identify complex haplotypes associated with AD. The authors identified 1127 significant AD-associated SNP triples, demonstrating that complex multi-SNP haplotypes—which may not include the canonical APOE ε4 or ε2 alleles—play definitive roles in AD predisposition, with the ε4 allele showing strengthened connections to other region alleles and ε2 showing weakened connections in affected subjects.

A candidate gene association study of dementia risk in 2,857 older postmenopausal women from the Women's Health Initiative Memory Study examining 96 SNPs across five genes (APOE/TOMM40, BDNF, COMT, SORL1, KIBRA). The APOE/TOMM40 locus showed the strongest association (rs157582: OR=1.64, p=2.4×10⁻⁸ for probable dementia), with additional significant associations in COMT (rs737865), BDNF (rs1491850), and KIBRA (rs4320284, rs2241368, rs244904). Results support APOE/TOMM40 as a dementia risk locus and extend associations to COMT, BDNF, and KIBRA genes.

Multi-cohort study investigating progranulin (GRN) mRNA expression in peripheral blood across Alzheimer's disease (AD), mild cognitive impairment (MCI), and control subjects. Progranulin expression was 13% higher in AD and MCI patients in the UCSF-MAC cohort (p=3.7×10⁻⁵) and replicated in AddNeuroMed (21% higher in AD, p=5.6×10⁻⁶), but not in ADNI. The rs5848 T-allele predicted 12.3% lower progranulin expression (p=0.02), and APOE4 showed independent positive association with expression.

This methods paper compares permutation feature importance (PFI) and built-in coefficients (BIC) for identifying important genetic variants in Random Forest models with non-additive epistatic interactions. Using simulated datasets and real data from Alzheimer's disease and Primary Open Angle Glaucoma studies, the authors show that PFI provides more precise feature importance ranking in the presence of gene-gene interactions.

This international doctoral thesis examined gene-physical activity interactions in fibromyalgia through six studies analyzing 64 SNPs across 34 candidate genes in Spanish women. The case-control study (314 fibromyalgia cases vs. 112 controls) identified associations of rs841 (GCH1), rs1799971 (OPRM1), and rs2097903 (COMT) with fibromyalgia susceptibility (p=0.04, p=0.02, and p=0.04 respectively). Cross-sectional studies (n=274-276 fibromyalgia patients) found that SCN9A rs4453709 and other genetic polymorphisms interacted with physical activity to influence pain, fatigue, and resilience outcomes.

This case-control study examined genetic variation in APP processing pathway genes and their association with cerebrospinal fluid amyloid-beta 42 levels in Parkinson's disease patients (n=85) versus healthy controls (n=161). Two SNPs showed significant correlation with CSF Aβ42 in PD: APP rs466448 (lower levels, p=0.014) and APH1B rs2068143 (higher levels, p=0.002), while three SNPs correlated in controls: APP rs214484 and rs2040273, and PSEN1 rs362344 (all lower levels). Results suggest APP and APH1B genetic variants may modulate CSF Aβ42 levels in PD patients.

This case report describes a 40-year-old African American man with homozygous apolipoprotein E (APOE) deficiency caused by a frameshift mutation (c.291delG, p.E97fs). Despite complete absence of apoE protein, the patient exhibited normal neurocognitive function, normal retinal function, normal brain MRI, and normal cerebrospinal fluid biomarkers (Aβ42=829 pg/mL, total tau=165 pg/mL, p-tau181=43.4 pg/mL). He had severe dysbetalipoproteinemia with total cholesterol of 760 mg/dL and multiple xanthomas, but no neurodegenerative disease.

A family-based genome-wide association study identified novel association between PLXNA4 SNPs (rs277470, meta-P=4.1×10⁻⁸) and Alzheimer disease across Framingham Heart Study and NIA-LOAD cohorts. Functional studies show the full-length PLXNA4 isoform (TS1) increases tau phosphorylation when stimulated by SEMA3A, while shorter isoforms have opposite effects. Brain tissue from late-stage AD cases showed 1.9-fold increased TS1 expression (P=1.6×10⁻⁴) correlated with dementia severity and neuropathology, supporting a role for PLXNA4 in tau-mediated AD pathogenesis.

Whole genome sequencing of 44 Ashkenazi Jewish centenarians identified 216 coding variants annotated as pathogenic or likely pathogenic in ClinVar. The study found 130 rare variants (MAF <5%) reported to cause degenerative, neoplastic, and cardiac diseases with various inheritance patterns. Notably, several carriers had no clinical manifestations despite carrying variants linked to serious diseases (e.g., an APOE ε4 homozygote without Alzheimer's disease, a UBQLN2 P525S carrier without ALS). These findings suggest incomplete penetrance and reduced clinical significance for many reported disease mutations.

This GWAS meta-analysis in 5,896 African Americans (1,968 cases, 3,928 controls) identified rs115550680 in ABCA7 as significantly associated with late-onset Alzheimer disease (AD) (OR=1.79, 95% CI 1.47-2.12, P=2.2×10⁻⁹), with effect size comparable to APOE ε4 (rs429358, OR=2.31, P=5.5×10⁻⁴⁷). Additional AD-associated loci identified in African Americans included variants in CR1, BIN1, EPHA1, and CD33, some differing from European ancestry findings.

This association study of 732 chronic hepatitis C patients examined polymorphisms in the TOMM40-APOE genomic region. rs7412 was significantly associated with serum apolipoprotein E levels (p=2.3×10⁻¹¹), explaining 7% of variance. Among IL28B-CC patients (n=196), rs429358 and TOMM40 '523' polymorphisms together explained 12% of variance in apolipoprotein B levels. Patients homozygous for APOE e3 isoform showed >2-fold increased odds of F2-F4 hepatic fibrosis (p=1.8×10⁻⁵), independent of lipid levels.

First genome-wide association study (GWAS) of HIV-associated neurocognitive disorders (HAND) in 1287 HIV-infected adults from the Multicenter AIDS Cohort Study. The study examined neurocognitive decline in processing speed and executive functioning over time, as well as prevalence of HIV-associated dementia and neurocognitive impairment across ~2.5 million SNPs. No genome-wide significant associations were identified with any neurocognitive phenotype examined. Previously reported candidate gene associations with HAND (including rs1130371 in MIP1α, rs1800629 in TNFα, rs1801157 in SDF-1, and rs2839619 in PREP1) were not validated in this study.

This candidate gene association study examined 18 SNPs in tau phosphorylation pathway genes (kinases and phosphatases) in relation to cerebrospinal fluid (CSF) tau levels in 101 Alzheimer's disease (AD) patients and 169 cognitively normal controls. Two SNPs significantly correlated with CSF tau levels after multiple comparison correction: rs7768046 in the FYN kinase gene (associated with increased CSF t-tau in AD, P=0.0007 Holm-corrected) and rs913275 in the PPP2R4 phosphatase gene (associated with increased CSF p-tau and t-tau in AD, P=0.0024 Holm-corrected). These findings suggest that genetic variation in genes regulating tau phosphorylation influences CSF tau levels in an AD-associated manner.

This dissertation study identified HLA region haplotypes associated with Alzheimer's disease risk using imputation and direct sequencing. In 11,690 combined UCSF and Alzheimer's Disease Genetics Consortium subjects, the five-allele haplotype A*03:01~B*07:02~DRB1*15:01~DQA1*01:02~DQB1*06:02 was associated with increased AD risk (OR=1.21, p=9.6×10⁻⁴). The DR15 haplotype (DRB1*15:01~DQA1*01:02~DQB1*06:02) was associated with faster cognitive decline and higher baseline inflammation. Direct sequencing confirmed these findings and revealed the six-allele haplotype A*03:01~B*07:02~C*07:02~DRB1*15:01~DQB1*06:02~DPB1*04:01 had a protective effect in atypical AD (OR=0.18, p=0.01) despite conferring risk in typical amnestic AD.

This case-control study attempted to replicate the association of TOMM40 polyT polymorphism (rs10524523) with Alzheimer's disease risk and age at onset. In a large series of 1594 LOAD cases and 1190 controls, the study failed to replicate the previously reported association with age at onset but found a significant association between rs10524523 and risk for LOAD among APOE 33 homozygotes in the opposite direction (OR=0.78, 95% CI=0.65-0.95; p=0.004, allele frequencies 0.41 vs 0.48 in cases vs controls). No association was found between rs10524523 and CSF biomarker levels (tau, phosphorylated tau, Aβ42) or TOMM40/APOE gene expression.

Pooled analysis of 15 studies (n=21,160) demonstrating that the APOE ε4 haplotype is protective against late age-related macular degeneration (AMD) (OR=0.72 per haplotype, p=4.41×10^-11), while the ε2 homozygote shows increased risk (OR=1.83, p=0.04). Extended haplotype analysis incorporating rs405509 did not identify additional risks beyond ε2 and ε4 haplotypes. Smoking was a major risk factor for progression to late AMD forms.

A genome-wide survival meta-analysis of 3,923 Parkinson's disease patients identified three genome-wide significant loci associated with progression to Parkinson's disease dementia: APOE rs429358 (HR=2.41, P=2.32×10−15), LRP1B rs80306347 (HR=3.23, P=7.07×10−9), and BBS9 rs78294974 (HR=3.90, P=3.59×10−8). The study reveals APOE ε4 and LRP1B as major risk factors and suggests the amyloid pathway's involvement in dementia development.

This 3-stage genome-wide association meta-analysis study identified 5 novel and confirmed loci associated with late-onset Alzheimer's disease across over 35,000 individuals. The study discovered two new genome-wide significant loci: rs744373 in BIN1 (OR 1.13, p=1.59×10⁻¹¹) and rs597668 near EXOC3L2 (OR 1.18, p=6.45×10⁻⁹), and confirmed three previously reported loci in APOE (OR 2.53, p=1.04×10⁻²⁹⁵), CLU (OR 0.85, p=1.62×10⁻¹⁶), and PICALM (OR 0.89, p=3.16×10⁻¹²). These findings were validated in an independent Spanish replication sample of 2,349 individuals.

This prospective cohort study of 593 older adults (≥70 years) examined associations between the CETP rs5882 (V405, valine/isoleucine) polymorphism and cognitive decline and dementia risk. Valine homozygotes showed 51% slower episodic memory decline compared to isoleucine homozygotes and had significantly lower risk for incident dementia (HR 0.28, 95% CI 0.10-0.85, P=0.02) and Alzheimer's disease (HR 0.31, 95% CI 0.10-0.95, P=0.03), suggesting a potential protective association of CETP valine homozygosity.

A nested case-control study of 309 postmenopausal osteoporotic women and 293 controls found that the IL6 -634G>C SNP (rs1800796) was associated with osteoporosis (OR 2.51, p=0.0047), independent of smoking or statin use. Additionally, the LRP5 C135242T SNP (rs545382) showed association with osteoporosis specifically in cigarette smokers (OR 2.8, p=0.03), suggesting a gene-environment interaction.

This cumulative PhD dissertation investigates genetic susceptibility factors for age-related macular degeneration (AMD). The study confirms weak associations of APOE coding variants with AMD risk (P < 0.05) but finds no association with HMCN1 variants. Large replication studies of candidate genes TLR3 and SERPING1 (1,080-4,881 cases and 2,669-2,842 controls) show no association. The authors identified 15 high-risk variants in ARMS2/HTRA1 region on chromosome 10q23.33-10qter, with the ARMS2 A69S variant showing 2.7-fold increased risk heterozygously and 8.2-fold increased risk homozygously, comparable in strength to CFH Y402H. An indel variant (c.*372_815del443ins54) in ARMS2 3' UTR causes mRNA destabilization.

Genome-wide linkage and association study of 1,902 individuals from 328 families with late-onset Alzheimer disease (LOAD) and 236 unrelated controls, using ~6,000 SNP markers. The strongest finding was at chromosome 19q13.32 confirming the APOE gene effect on LOAD risk (FBAT Z=8.68, P=1.98×10⁻¹⁸; case-control χ²=150.46, P=1.4×10⁻³⁴). Additional significant loci identified include 7p22.2 (rs798485, LOD=3.77), 7p21.3 (rs719423, LOD=2.89), and 16q21 (rs1482258, LOD=3.32) in linkage analyses, with 7q31.1 and 20q13.33 also showing positive associations in case-control and meta-analysis comparisons.

An association study of 201 warfarin-treated patients found that polymorphisms in VKORC1, CYP2C9, CYP2C18, CYP2C19, PROC, and APOE were significantly associated with warfarin dose requirement (P < 0.000175 for VKORC1 and CYP2C9). A multiple regression model incorporating VKORC1, CYP2C9, PROC, and non-genetic factors (age, bodyweight, drug interactions, treatment indication) accounted for 62% of the variance in warfarin dose.

This dissertation study identified HLA haplotypes associated with Alzheimer's disease risk using case-control imputation and direct sequencing. The haplotype A*03:01~B*07:02~DRB1*15:01~DQA1*01:02~DQB1*06:02 was associated with increased AD risk (p=9.6×10⁻⁴, OR=1.21 [1.08-1.37]) in 11,690 combined UCSF and ADGC participants, with effects primarily in APOE4 non-carriers. A separate haplotype showed decreased risk for atypical AD (p=0.01, OR=0.18). The findings highlight the role of immune-related genetic variation in AD pathophysiology.

This study examined 2,861 individuals from three European populations to assess the association between ACE gene variants and age-at-onset (AAO) of Alzheimer's disease. Three ACE SNPs (rs4343, rs4291, rs1800764) were genotyped; rs4343 showed strong association with AAO (P<0.0001) with the GG genotype associated with later disease onset, and rs4291 showed independent effects (P=0.0095). The effects were consistent across gender and APOE ε4 carrier status.

This doctoral thesis conducted multiple candidate gene association studies in 274-426 southern Spanish women with fibromyalgia to investigate gene-physical activity/sedentary behavior interactions with pain, fatigue, and resilience. Study III identified rs841 (GCH1) GG genotype (OR=0.61, p=0.04) and rs2097903 (COMT) AT/TT genotypes (OR=1.66, p=0.04) associated with fibromyalgia susceptibility, and confirmed rs1799971 (OPRM1) GG genotype (OR=0.58, p=0.02) confers genetic risk. Study IV found rs6311/rs6313 (HTR2A) polymorphisms individually associated with algometer pain score, and gene-sedentary behavior interactions involving rs4680/rs165599 (COMT), rs1383914 (ADRA1A), rs12994338/rs4453709 (SCN9A), and rs6860 (CHMP1A) significantly associated with pain outcomes. SCN9A emerged as most robust gene for fibromyalgia phenotype.