Publications

@article{Han:HumMolGenet:2016,
title = {A general framework for meta-analyzing dependent studies with overlapping subjects in association mapping.},
author = {Buhm Han and Dat Duong and Jae Hoon Sul and Paul I. W. de Bakker and Eleazar Eskin and Soumya Raychaudhuri},
url = {http://dx.doi.org/10.1093/hmg/ddw049},
issn = {1460-2083},
year = {2016},
date = {2016-01-01},
journal = {Hum Mol Genet},
abstract = {Meta-analysis strategies have become critical to augment power of genome-wide association studies (GWAS). To reduce genotyping or sequencing cost, many studies today utilize shared controls, and these individuals can inadvertently overlap among multiple studies. If these overlapping individuals are not taken into account in meta-analysis, they can induce spurious associations. In this paper, we propose a general framework for adjusting association statistics to account for overlapping subjects within a meta-analysis. The key idea of our method is to transform the covariance structure of the data so it can be used in downstream analyses. As a result, the strategy is very flexible, and allows a wide range of meta-analysis methods, such as the random effects model, to account for overlapping subjects. Using simulations and real datasets, we demonstrate that our method has utility in meta-analyses of GWAS, as well as in a multi-tissue mouse eQTL study where our method increases the number of discovered eQTLs by up to 19% compared to existing methods},
keywords = {eQTL, genome-wide association studies, Meta-Analysis},
pubstate = {published},
tppubtype = {article}
}

@article{Sul:PlosGenet:2016,
title = {Accounting for Population Structure in Gene-by-Environment Interactions in Genome-Wide Association Studies Using Mixed Models.},
author = {Jae Hoon Sul and Michael Bilow and Wen-Yun Y. Yang and Emrah Kostem and Nick Furlotte and Dan He and Eleazar Eskin},
url = {http://dx.doi.org/10.1371/journal.pgen.1005849},
issn = {1553-7404},
year = {2016},
date = {2016-01-01},
journal = {PLoS Genet},
volume = {12},
number = {3},
pages = {e1005849},
address = {United States},
abstract = {Although genome-wide association studies (GWASs) have discovered numerous novel genetic variants associated with many complex traits and diseases, those genetic variants typically explain only a small fraction of phenotypic variance. Factors that account for phenotypic variance include environmental factors and gene-by-environment interactions (GEIs). Recently, several studies have conducted genome-wide gene-by-environment association analyses and demonstrated important roles of GEIs in complex traits. One of the main challenges in these association studies is to control effects of population structure that may cause spurious associations. Many studies have analyzed how population structure influences statistics of genetic variants and developed several statistical approaches to correct for population structure. However, the impact of population structure on GEI statistics in GWASs has not been extensively studied and nor have there been methods designed to correct for population structure on GEI statistics. In this paper, we show both analytically and empirically that population structure may cause spurious GEIs and use both simulation and two GWAS datasets to support our finding. We propose a statistical approach based on mixed models to account for population structure on GEI statistics. We find that our approach effectively controls population structure on statistics for GEIs as well as for genetic variants},
keywords = {gene-by-environment interactions, genome-wide association studies, Mixed Models},
pubstate = {published},
tppubtype = {article}
}

@article{Joo:GenomeBiol:2016,
title = {Multiple testing correction in linear mixed models.},
author = {Jong Wha J. Joo and Farhad Hormozdiari and Buhm Han and Eleazar Eskin},
url = {http://dx.doi.org/10.1186/s13059-016-0903-6},
issn = {1474-760X},
year = {2016},
date = {2016-01-01},
journal = {Genome Biol},
volume = {17},
number = {1},
pages = {62},
address = {England},
abstract = {BACKGROUND: Multiple hypothesis testing is a major issue in genome-wide association studies (GWAS), which often analyze millions of markers. The permutation test is considered to be the gold standard in multiple testing correction as it accurately takes into account the correlation structure of the genome. Recently, the linear mixed model (LMM) has become the standard practice in GWAS, addressing issues of population structure and insufficient power. However, none of the current multiple testing approaches are applicable to LMM. RESULTS: We were able to estimate per-marker thresholds as accurately as the gold standard approach in real and simulated datasets, while reducing the time required from months to hours. We applied our approach to mouse, yeast, and human datasets to demonstrate the accuracy and efficiency of our approach. CONCLUSIONS: We provide an efficient and accurate multiple testing correction approach for linear mixed models. We further provide an intuition about the relationships between per-marker threshold, genetic relatedness, and heritability, based on our observations in real data},
keywords = {genome-wide association studies, Mixed Models, Multiple Testing},
pubstate = {published},
tppubtype = {article}
}

@article{Lusis:JLipidRes:2016,
title = {The Hybrid Mouse Diversity Panel: A Resource for Systems Genetics Analyses of Metabolic and Cardiovascular Traits.},
author = {Aldons J. Lusis and Marcus Seldin and Hooman Allayee and Brian J. Bennett and Mete Civelek and Richard C. Davis and Eleazar Eskin and Charles Farber and Simon T. Hui and Margarete Mehrabian and Frode Norheim and Calvin Pan and Brian Parks and Christoph Rau and Desmond J. Smith and Thomas Vallim and Yibin Wang and Jessica Wang},
url = {http://dx.doi.org/10.1194/jlr.R066944},
issn = {1539-7262},
year = {2016},
date = {2016-01-01},
journal = {J Lipid Res},
abstract = {The Hybrid Mouse Diversity Panel (HMDP) is a collection of approximately 100 well-characterized inbred strains of mice that can be used to analyze the genetic and environmental factors underlying complex traits. While not nearly as powerful for mapping genetic loci contributing to the traits as human Genome-Wide Association Studies (GWAS), it has some important advantages. First, environmental factors can be controlled. Second, relevant tissues are accessible for global molecular phenotyping. Finally, because inbred strains are renewable, results from separate studies can be integrated. Thus far, the HMDP has been studied for traits relevant to obesity, diabetes, atherosclerosis, osteoporosis, heart failure, immune regulation, fatty liver disease, and host-gut microbiota interactions. High-throughput technologies have been used to examine the genomes, epigenomes, transcriptomes, proteomes, metabolomes, and microbiomes of the mice under various environmental conditions. All of the published data are available and can be readily used to formulate hypotheses about genes, pathways and interactions},
keywords = {genome-wide association studies, Hybrid Mouse Diversity Panel, Mouse Genetics},
pubstate = {published},
tppubtype = {article}
}

@article{Lavinsky:PlosGenet:2015,
title = {Genome-wide association study identifies nox3 as a critical gene for susceptibility to noise-induced hearing loss.},
author = { Joel Lavinsky and Amanda L. Crow and Calvin Pan and Juemei Wang and Ksenia A. Aaron and Maria K. Ho and Qingzhong Li and Pehzman Salehide and Anthony Myint and Maya Monges-Hernadez and Eleazar Eskin and Hooman Allayee and Aldons J. Lusis and Rick A. Friedman},
url = {http://dx.doi.org/10.1371/journal.pgen.1005293},
issn = {1553-7404},
year = {2015},
date = {2015-01-01},
volume = {11},
number = {6},
pages = {e1005293},
address = {United States},
keywords = {cochlear function, genome-wide association studies},
pubstate = {published},
tppubtype = {article}
}

@inbook{Joo:ResearchInComputationalMolecularBiology:2015b,
title = {Efficient and Accurate Multiple-Phenotypes Regression Method for High Dimensional Data Considering Population Structure},
author = {Jong Wha J. Joo and Eun Yong Kang and Elin Org and Nick Furlotte and Brian Parks and Aldons J. Lusis and Eleazar Eskin},
url = {http://dx.doi.org/10.1007/978-3-319-16706-0_15},
year = {2015},
date = {2015-01-01},
booktitle = {Research in Computational Molecular Biology},
pages = {136-153},
publisher = {Springer International Publishing},
organization = {University of California},
abstract = {A typical GWAS tests correlation between a single phenotype and each genotype one at a time. However, it is often very useful to analyze many phenotypes simultaneously. For example, this may increase the power to detect variants by capturing unmeasured aspects of complex biological networks that a single phenotype might miss. There are several multivariate approaches that try to detect variants related to many phenotypes, but none of them consider population structure and each may result in a significant number of false positive identifications. Here, we introduce a new methodology, referred to as GAMMA, that could both simultaneously analyze many phenotypes as well as correct for population structure. In a simulated study, GAMMA accurately identifies true genetic effects without false positive identifications, while other methods either fail to detect true effects or result in many false positive identifications. We further apply our method to genetic studies of yeast and gut microbiome from mouse and show that GAMMA identifies several variants that are likely to have a true biological mechanism.},
keywords = {GAMMA, genome-wide association studies, microbiome},
pubstate = {published},
tppubtype = {inbook}
}

@article{Wang:JComputBiol:2015b,
title = {Gene-Gene Interactions Detection Using a Two-stage Model.},
author = { Zhanyong Wang and Jae Hoon Sul and Sagi Snir and Jose A. Lozano and Eleazar Eskin},
url = {http://dx.doi.org/10.1089/cmb.2014.0163},
issn = {1557-8666},
year = {2015},
date = {2015-01-01},
journal = {J Comput Biol},
volume = {22},
number = {6},
pages = {563-76},
address = {United States},
abstract = {Genome-wide association studies (GWAS) have discovered numerous loci involved in genetic traits. Virtually all studies have reported associations between individual single nucleotide polymorphisms (SNPs) and traits. However, it is likely that complex traits are influenced by interaction of multiple SNPs. One approach to detect interactions of SNPs is the brute force approach which performs a pairwise association test between a trait and each pair of SNPs. The brute force approach is often computationally infeasible because of the large number of SNPs collected in current GWAS studies. We propose a two-stage model, Threshold-based Efficient Pairwise Association Approach (TEPAA), to reduce the number of tests needed while maintaining almost identical power to the brute force approach. In the first stage, our method performs the single marker test on all SNPs and selects a subset of SNPs that achieve a certain significance threshold. In the second stage, we perform a pairwise association test between traits and pairs of the SNPs selected from the first stage. The key insight of our approach is that we derive the joint distribution between the association statistics of a single SNP and the association statistics of pairs of SNPs. This joint distribution allows us to provide guarantees that the statistical power of our approach will closely approximate the brute force approach. We applied our approach to the Northern Finland Birth Cohort data and achieved 63 times speedup while maintaining 99% of the power of the brute force approach},
keywords = {genome-wide association studies, Threshold-based Efficient Pairwise Association Approach},
pubstate = {published},
tppubtype = {article}
}

@article{Luykx:MolPsychiatry:2015b,
title = {Genome-wide association study of NMDA receptor coagonists in human cerebrospinal fluid and plasma.},
author = { J. J. Luykx and S. C. Bakker and W. F. Visser and N. Verhoeven-Duif and J. E. Buizer-Voskamp and J. M. den Heijer and M. P. M. Boks and J. H. Sul and E. Eskin and A. P. Ori and R. M. Cantor and J. Vorstman and E. Strengman and J. DeYoung and T. H. Kappen and E. Pariama and E. P. A. van Dongen and P. Borgdorff and P. Bruins and T. J. de Koning and R. S. Kahn and R. A. Ophoff},
url = {http://dx.doi.org/10.1038/mp.2014.190},
issn = {1476-5578},
year = {2015},
date = {2015-01-01},
journal = {Mol Psychiatry},
abstract = {The N-methyl-d-aspartate receptor (NMDAR) coagonists glycine, d-serine and l-proline play crucial roles in NMDAR-dependent neurotransmission and are associated with a range of neuropsychiatric disorders. We conducted the first genome-wide association study of concentrations of these coagonists and their enantiomers in plasma and cerebrospinal fluid (CSF) of human subjects from the general population (N=414). Genetic variants at chromosome 22q11.2, located in and near PRODH (proline dehydrogenase), were associated with l-proline in plasma ($beta$=0.29; P=6.38 $times$ 10(-10)). The missense variant rs17279437 in the proline transporter SLC6A20 was associated with l-proline in CSF ($beta$=0.28; P=9.68 $times$ 10(-9)). Suggestive evidence of association was found for the d-serine plasma-CSF ratio at the d-amino-acid oxidase (DAO) gene ($beta$=-0.28; P=9.08 $times$ 10(-8)), whereas a variant in SRR (that encodes serine racemase and is associated with schizophrenia) constituted the most strongly associated locus for the l-serine to d-serine ratio in CSF. All these genes are highly expressed in rodent meninges and choroid plexus, anatomical regions relevant to CSF physiology. The enzymes and transporters they encode may be targeted to further construe the nature of NMDAR coagonist involvement in NMDAR gating. Furthermore, the highlighted genetic variants may be followed up in clinical populations, for example, schizophrenia and 22q11 deletion syndrome. Overall, this targeted metabolomics approach furthers the understanding of NMDAR coagonist concentration variability and sets the stage for non-targeted CSF metabolomics projects.Molecular Psychiatry advance online publication, 10 February 2015; doi:10.1038/mp.2014.190},
keywords = {genome-wide association studies, NMDAR, schizophrenia},
pubstate = {published},
tppubtype = {article}
}

@article{Crow:G3:2015b,
title = {The Genetic Architecture of Hearing Impairment in Mice: Evidence for Frequency Specific Genetic Determinants.},
author = { Amanda L. Crow and Jeffrey Ohmen and Juemei Wang and Joel Lavinsky and Jaana Hartiala and Qingzhong Li and Xin Li and Pezhman Salehide and Eleazar Eskin and Calvin Pan and Aldons J. Lusis and Hooman Allayee and Rick A. Friedman},
url = {http://dx.doi.org/10.1534/g3.115.021592},
issn = {2160-1836},
year = {2015},
date = {2015-01-01},
journal = {G3 (Bethesda)},
abstract = {Genome-wide association studies (GWAS) have been successfully applied in humans for the study of many complex phenotypes. However, identification of the genetic determinants of hearing in adults has been hampered, in part, by the relative inability to control for environmental factors that might affect hearing throughout the lifetime, as well as a large degree of phenotypic heterogeneity. These and other factors have limited the number of large-scale studies performed in humans that have identified candidate genes that contribute to the etiology of this complex trait. In order to address these limitations, we performed a GWAS analysis using a set of inbred mouse strains from the Hybrid Mouse Diversity Panel. Among 99 strains characterized, we observed ~2 to 5-fold variation in hearing at six different frequencies, which are differentiated biologically from each other by the location in the cochlea where each frequency is registered. Among all frequencies tested, we identified a total of nine significant loci, several of which contained promising candidate genes for follow-up study. Taken together, our results indicate the existence of both genes that affect global cochlear function, as well as anatomical- - and frequency-specific genes, and further demonstrate the complex nature of mammalian hearing variation},
keywords = {cochlear function, genome-wide association studies, phenotypic heterogeneity},
pubstate = {published},
tppubtype = {article}
}

@article{Bennett:PlosGenet:2015,
title = {Genetic Architecture of Atherosclerosis in Mice: A Systems Genetics Analysis of Common Inbred Strains.},
author = {Brian J. Bennett and Richard C. Davis and Mete Civelek and Luz Orozco and Judy Wu and Hannah Qi and Calvin Pan and René R. Sevag Packard and Eleazar Eskin and Mujing Yan and Todd Kirchgessner and Zeneng Wang and Xinmin Li and Jill C. Gregory and Stanley L. Hazen and Peter S. Gargalovic and Aldons JLusis},
url = {http://dx.doi.org/10.1371/journal.pgen.1005711},
issn = {1553-7404},
year = {2015},
date = {2015-01-01},
journal = {PLoS Genet},
volume = {11},
number = {12},
pages = {e1005711},
address = {United States},
abstract = {Common forms of atherosclerosis involve multiple genetic and environmental factors. While human genome-wide association studies have identified numerous loci contributing to coronary artery disease and its risk factors, these studies are unable to control environmental factors or examine detailed molecular traits in relevant tissues. We now report a study of natural variations contributing to atherosclerosis and related traits in over 100 inbred strains of mice from the Hybrid Mouse Diversity Panel (HMDP). The mice were made hyperlipidemic by transgenic expression of human apolipoprotein E-Leiden (APOE-Leiden) and human cholesteryl ester transfer protein (CETP). The mice were examined for lesion size and morphology as well as plasma lipid, insulin and glucose levels, and blood cell profiles. A subset of mice was studied for plasma levels of metabolites and cytokines. We also measured global transcript levels in aorta and liver. Finally, the uptake of acetylated LDL by macrophages from HMDP mice was quantitatively examined. Loci contributing to the traits were mapped using association analysis, and relationships among traits were examined using correlation and statistical modeling. A number of conclusions emerged. First, relationships among atherosclerosis and the risk factors in mice resemble those found in humans. Second, a number of trait-loci were identified, including some overlapping with previous human and mouse studies. Third, gene expression data enabled enrichment analysis of pathways contributing to atherosclerosis and prioritization of candidate genes at associated loci in both mice and humans. Fourth, the data provided a number of mechanistic inferences; for example, we detected no association between macrophage uptake of acetylated LDL and atherosclerosis. Fifth, broad sense heritability for atherosclerosis was much larger than narrow sense heritability, indicating an important role for gene-by-gene interactions. Sixth, stepwise linear regression showed that the combined variations in plasma metabolites, including LDL/VLDL-cholesterol, trimethylamine N-oxide (TMAO), arginine, glucose and insulin, account for approximately 30 to 40% of the variation in atherosclerotic lesion area. Overall, our data provide a rich resource for studies of complex interactions underlying atherosclerosis},
keywords = {atherosclerosis, genome-wide association studies, HMDP},
pubstate = {published},
tppubtype = {article}
}