Common Contracts
2 similar Research Agreement contracts
AGREEMENT AMONG GWAS RESULTS FROM DIFFERENT STATISTICAL METHODS AS A STRATEGY TO INCREASE THE POWER OF QTL DETECTIONResearch Agreement • November 6th, 2019
The power of true positive associations in GWAS for traits affected by many QTL is generally low. This and other unfavorable scenarios pose a problem for the detection of true QTLs, which may lead to false positive associations. The aim of this study was to evaluate if combining the results of different statistical methods may increase the power to detect QTL. We simulated a polygenic trait, with known QTL positions. GWAS was performed using the WssGBLUP and BayesC methods, in a total of 8 different analyses, varying the assumptions of the SNP effects and the phenotypic data used. The results showed that as the number of analyses that a window was detected as important increased, so did the probability of that window containing a true QTL. Windows identified in 7 or 8 analyses were able to detect just some (60.5%) of the true QTL. Windows detected in at least 5 analyses captured 96% of the true QTL, but included some false positives (10.8%). Further studies are recommended, simulating
AGREEMENT AMONG GWAS RESULTS FROM DIFFERENT STATISTICAL METHODS AS A STRATEGY TO INCREASE THE POWER OF QTL DETECTIONResearch Agreement • December 14th, 2017
The power of true positive associations in GWAS for traits affected by many QTL is generally low. This and other unfavorable scenarios pose a problem for the detection of true QTLs, which may lead to false positive associations. The aim of this study was to evaluate if combining the results of different statistical methods may increase the power to detect QTL. We simulated a polygenic trait, with known QTL positions. GWAS was performed using the WssGBLUP and BayesC methods, in a total of 8 different analyses, varying the assumptions of the SNP effects and the phenotypic data used. The results showed that as the number of analyses that a window was detected as important increased, so did the probability of that window containing a true QTL. Windows identified in 7 or 8 analyses were able to detect just some (60.5%) of the true QTL. Windows detected in at least 5 analyses captured 96% of the true QTL, but included some false positives (10.8%). Further studies are recommended, simulating
