| Summary: | Abstract Background: Genetic association studies have been increasingly used in cattle breeding programs. However, inconsistent results -such as positive, negative, or absence of association- across studies restrain reproducibility and proper implementation, propitiating the occurrence of bias. Objective: To identify and classify potential sources of bias and determine possible strategies to avoid it in genetic association studies in cattle. Source of bias in genetic association studies: Genetic and genomic sources of bias include effects associated with the gene loci governing expression. Sampling-related and statistical biases are related with factors such as stratification and database size. Strategies to correct bias in genetic association studies: Correction strategies differ in nature. Genetic and genomic strategies are based on determining the appropriate approach to obtain and report the genetic information. Sampling-related and statistical strategies are based on grouping individuals with certain traits that lead to a reduction in heterogeneity. Conclusion: It is necessary to consider the methodology used in previous studies to establish a hierarchy of sources of bias and facilitate decisions on the use of tools to reduce inconsistencies in the results of future studies.
|
|---|
