Development, Preparation, and Curation of High-Throughput Phenotypic Data for Genome-Wide Association Studies: A Sample Pipeline in R

Abstract

Genome-wide association studies (GWAS) have benefited from the advances of sequencing methods for the generation of high-density genomic data. By bridging genotype to phenotype, several genes have been associated with traits of agricultural interest. Despite this, there is still a gap between genotyping and phenotyping due to the large difference in throughput between the two disciplines. Although cutting-edge phenomics technologies are available to the community, their costs are still prohibitive at the small lab level. Semiautomated methods of investigation provide a valid alternative to generate large-scale phenotyping data able to deeply investigate the characteristics of different plant organs. Beyond automation, phenomics data management is another major constraint to consider; while bioinformatics pipelines are well-trained for releasing high-quality genomic data, fewer efforts have been done for phenotyping information. This chapter provides a guide for generating large-scale data related to the size and shape of fruits, leaves, seeds, and roots and for downstream analysis for curation and preparation of clean datasets, through removal of outliers and performing primary statistical analysis. Different steps to be carried out in the R environment will be shown for gathering the appropriate input information to use in GWAS avoiding any possible bias.