GAPIT Version 3: Boosting Power and Accuracy for Genomic Association and Prediction

Area of Science:

• Genomics
• Bioinformatics
• Statistical Genetics

Background:

• Genome-wide association study (GWAS) and genomic prediction/selection (GP/GS) are crucial in genomic research.
• Advancements in analytical methods and software are necessary due to the complexity of genomic and phenotypic data.
• GAPIT is a widely-used R package for integrated genomic association and prediction.

Purpose of the Study:

• To document the upgrades in GAPIT version 3, focusing on new multi-locus GWAS and GP/GS methods.
• To introduce novel implementations that enhance statistical power, prediction accuracy, and computational efficiency.
• To provide accessible resources for researchers utilizing GAPIT.

Main Methods:

• Implementation of three multi-locus GWAS methods: multiple loci mixed model (MLMM), fixed and random model circulating probability unification (FarmCPU), and Bayesian-information and linkage-disequilibrium iteratively nested keyway (BLINK).
• Introduction of two GP/GS methods: compressed BLUP (cBLUP) based on CMLM and SUPER BLUP (sBLUP) based on SUPER.
• Leveraging existing GAPIT framework for integrated analysis of genomic and phenotypic data.

Main Results:

• GAPIT version 3 significantly boosts statistical power for GWAS through multi-locus approaches.
• Prediction accuracy for GP/GS is improved with the new cBLUP and sBLUP methods.
• Enhanced computational speed and capacity for analyzing large-scale genomic datasets.

Conclusions:

• GAPIT version 3 represents a substantial advancement in genomic analysis tools.
• The new multi-locus methods and GP/GS algorithms offer improved performance and efficiency.
• Freely available resources facilitate the adoption and application of these advanced genomic techniques.