TY - GEN
T1 - Benefits of accurate imputations in GWAS
AU - Verma, Shefali S.
AU - Peissig, Peggy
AU - Cross, Deanna
AU - Waudby, Carol
AU - Brilliant, Murray
AU - McCarty, Catherine A.
AU - Ritchie, Marylyn D.
N1 - Publisher Copyright:
© Springer-Verlag Berlin Heidelberg 2014.
PY - 2014
Y1 - 2014
N2 - 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.
AB - 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.
UR - http://www.scopus.com/inward/record.url?scp=84915818731&partnerID=8YFLogxK
UR - http://www.scopus.com/inward/citedby.url?scp=84915818731&partnerID=8YFLogxK
U2 - 10.1007/978-3-662-45523-4_71
DO - 10.1007/978-3-662-45523-4_71
M3 - Conference contribution
AN - SCOPUS:84915818731
T3 - Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics)
SP - 877
EP - 889
BT - Applications of Evolutionary Computation - 17th European Conference, EvoApplications 2014, Revised Selected Papers
A2 - Esparcia-Alcázar, Anna I.
PB - Springer Verlag
T2 - 17th European Conference on Applications of Evolutionary Computation, EvoApplications 2014
Y2 - 23 April 2014 through 25 April 2014
ER -