Dataset Information

Evaluation of a phenotype imputation approach using GAW20 simulated data.

ABSTRACT: Statistical power, which is the probability of correctly rejecting a false null hypothesis, is a limitation of genome-wide association studies (GWAS). Sample size is a major component of statistical power that can be easily affected by missingness in phenotypic data and restrain the ability to detect associated single-nucleotide polymorphisms (SNPs) with small effect sizes. Although some phenotypes are hard to collect because of cost and loss to follow-up, correlated phenotypes that are easily collected can be leveraged for association analysis. In this paper, we evaluate a phenotype imputation method that incorporates family structure and correlation between multiple phenotypes using GAW20 simulated data. The distribution of missing values is derived using information contained in the missing sample's relatives and additional correlated phenotypes. We show that this imputation method can improve power in the association analysis compared with excluding observations with missing data, while achieving the correct Type I error rate. We also examine factors that may affect the imputation accuracy.

SUBMITTER: Chen Y

PROVIDER: S-EPMC6157247 | biostudies-literature | 2018

REPOSITORIES: biostudies-literature

ACCESS DATA

Publications

Evaluation of a phenotype imputation approach using GAW20 simulated data.

Chen Yuning Y Peloso Gina M GM Dupuis Josée J

BMC proceedings 20180917 Suppl 9

Statistical power, which is the probability of correctly rejecting a false null hypothesis, is a limitation of genome-wide association studies (GWAS). Sample size is a major component of statistical power that can be easily affected by missingness in phenotypic data and restrain the ability to detect associated single-nucleotide polymorphisms (SNPs) with small effect sizes. Although some phenotypes are hard to collect because of cost and loss to follow-up, correlated phenotypes that are easily c ...[more]

PMID: 30275899

Similar Datasets

Project description:BackgroundIncomplete categorical variables with more than two categories are common in public health data. However, most of the existing missing-data methods do not use the information from nonresponse (missingness) probabilities.MethodsWe propose a nearest-neighbour multiple imputation approach to impute a missing at random categorical outcome and to estimate the proportion of each category. The donor set for imputation is formed by measuring distances between each missing value with other non-missing values. The distance function is calculated based on a predictive score, which is derived from two working models: one fits a multinomial logistic regression for predicting the missing categorical outcome (the outcome model) and the other fits a logistic regression for predicting missingness probabilities (the missingness model). A weighting scheme is used to accommodate contributions from two working models when generating the predictive score. A missing value is imputed by randomly selecting one of the non-missing values with the smallest distances. We conduct a simulation to evaluate the performance of the proposed method and compare it with several alternative methods. A real-data application is also presented.ResultsThe simulation study suggests that the proposed method performs well when missingness probabilities are not extreme under some misspecifications of the working models. However, the calibration estimator, which is also based on two working models, can be highly unstable when missingness probabilities for some observations are extremely high. In this scenario, the proposed method produces more stable and better estimates. In addition, proper weights need to be chosen to balance the contributions from the two working models and achieve optimal results for the proposed method.ConclusionsWe conclude that the proposed multiple imputation method is a reasonable approach to dealing with missing categorical outcome data with more than two levels for assessing the distribution of the outcome. In terms of the choices for the working models, we suggest a multinomial logistic regression for predicting the missing outcome and a binary logistic regression for predicting the missingness probability.

Dataset Information

Evaluation of a phenotype imputation approach using GAW20 simulated data.

Publications

Evaluation of a phenotype imputation approach using GAW20 simulated data.

Similar Datasets

OmicsDI is part of the ELIXIR infrastructure

Tweets