Genome-wide epistasis analysis for Alzheimer’s disease and implications for genetic risk prediction

Abstract

Background Single-nucleotide polymorphisms (SNPs) identified by genome-wide association studies only explain part of the heritability of Alzheimer’s disease (AD). Epistasis has been considered as one of the main causes of “missing heritability” in AD. Methods We performed genome-wide epistasis screening (N = 10,389) for the clinical diagnosis of AD using three popularly adopted methods. Subsequent analyses were performed to eliminate spurious associations caused by possible confounding factors. Then, candidate genetic interactions were examined for their co-expression in the brains of AD patients and analyzed for their association with intermediate AD phenotypes. Moreover, a new approach was developed to compile the epistasis risk factors into an epistasis risk score (ERS) based on multifactor dimensional reduction. Two independent datasets were used to evaluate the feasibility of ERSs in AD risk prediction. Results We identified 2 candidate genetic interactions with PFDR < 0.05 (RAMP3-SEMA3A and NSMCE1-DGKE/C17orf67) and another 5 genetic interactions with PFDR < 0.1. Co-expression between the identified interactions supported the existence of possible biological interactions underlying the observed statistical significance. Further association of candidate interactions with intermediate phenotypes helps explain the mechanisms of neuropathological alterations involved in AD. Importantly, we found that ERSs can identify high-risk individuals showing earlier onset of AD. Combined risk scores of SNPs and SNP-SNP interactions showed slightly but steadily increased AUC in predicting the clinical status of AD. Conclusions In summary, we performed a genome-wide epistasis analysis to identify novel genetic interactions potentially implicated in AD. We found that ERS can serve as an indicator of the genetic risk of AD.

Measuring heritable contributions to Alzheimer’s disease: polygenic risk score analysis with twins

“… 42 , 43 The missing heritability is likely not due to simple additivity across common variants but also to contributions from rare variants as well as to non-additive effects including dominance and epistasis. 42 , 44 Studies of rare variants and Alzheimer’s disease risk have observed effects for rare coding variants in genes such as ABCA7, BIN, NOTCH3, PLCG2, SORL1, TREM and ABI3 among others 45–47 not captured by PRSs. Apart from a rare variant in TREM2 (p.Arg47His), little replication work has been reported.…”

Systems Genetic Identification of Mitochondrion-Associated Alzheimer’s Disease Genes and Implications for Disease Risk Prediction

“…In addition, the predictive power of the hub genes was compared between ADNI DEGs and the recognized AD pathogenic genes [ 2 ] ( Table S12 ). APOE is the most important genetic factor of AD and accounts for approximately 5–9% of heritability [ 39 ]; thus, the prediction efficiency of hub genes was compared with recognized pathogenic genes other than APOE . ROC curves showed that the AUC of hub genes (Hub Genes AUC: 0.623) was significantly higher than that of AD pathogenic genes and AD pathogenic genes excluding APOE (Sig Genes AUC: 0.541, Sig~ APOE AUC: 0.528; DeLong’s test, p = 0.02/0.008) ( Figure 4 A).…”

Ultra-marginal Feature Importance

“…Interaction effects are common in many scientific disciplines where assessing feature importance is prevalent including hydrology [31,32,33], genomics [4,34,35], and finance [36,37]. So, as was done in [7], we assess the ability of MCI and UMFI to detect non-linear interaction effects in the data.…”