Benefits of accurate imputations in GWAS

Shefali S. Verma, Peggy Peissig, Deanna Cross, Carol Waudby, Murray Brilliant, Catherine A. McCarty, Marylyn D. Ritchie

Research output: Chapter in Book/Report/Conference proceedingConference contribution

1 Scopus citations


Imputation methods have been suggested as an efficient way to increase both utility and coverage in genome-wide association studies, especially when combining data generated from different genotyping arrays. We aim to demonstrate that imputation results are extremely accurate and the association analysis from imputed data does not over-inflate the results. Instead imputation leads to an increase in the power of the dataset without introducing any systematic biases. The majority of common variants can be imputed with very high accuracy (r2>0.9) and we validated the accuracy of imputations by comparing actual genotypes from low-throughput genotyping assays against imputed genotypes. Imputation was performed using IMPUTE2 and the 1000 Genomes cosmopolitan reference panel, which results in about 38 million SNPs. After quality control and filtering we performed case-control associations with 3,159,556 markers. We show a comparison of results from genotyped and imputed data and also determine how accurate ancestry is determined by imputations.

Original languageEnglish (US)
Title of host publicationApplications of Evolutionary Computation - 17th European Conference, EvoApplications 2014, Revised Selected Papers
EditorsAnna I. Esparcia-Alcázar
PublisherSpringer Verlag
Number of pages13
ISBN (Electronic)9783662455227
StatePublished - 2014
Event17th European Conference on Applications of Evolutionary Computation, EvoApplications 2014 - Granada, Spain
Duration: Apr 23 2014Apr 25 2014

Publication series

NameLecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics)
ISSN (Print)0302-9743
ISSN (Electronic)1611-3349


Other17th European Conference on Applications of Evolutionary Computation, EvoApplications 2014

All Science Journal Classification (ASJC) codes

  • Theoretical Computer Science
  • General Computer Science


Dive into the research topics of 'Benefits of accurate imputations in GWAS'. Together they form a unique fingerprint.

Cite this