GCA

GCA computes genetic connectedness statistics from pedigree and genomic data to assess comparability of predicted genetic values across management units and between training and testing sets in whole-genome prediction.

Key Features:

• Input data types: Accepts pedigree and genomic data as the basis for connectedness analysis.
• Connectedness statistics suite: Implements a comprehensive set of connectedness statistics for quantifying linkage between units and individuals.
• Statistical basis: Calculates metrics as functions of prediction error variance (PEV) or the variance of unit effect estimates (VE).
• Training–testing assessment: Measures the relationship between training and testing sets to inform whole-genome prediction comparisons.
• Unit-level linkage quantification: Quantifies genetic connectedness among individuals within and between management units.
• Support for genetic evaluation: Provides metrics to compare predicted genetic values across different management units for genomic prediction and evaluation.

Scientific Applications:

• Comparability assessment: Assess the comparability of predicted genetic values across management units.
• Genomic prediction validation: Evaluate connectedness between training and testing sets in whole-genome prediction to gauge prediction reliability.
• Genetic evaluation support: Inform decisions in genetic evaluation and genomic prediction by quantifying linkage and uncertainty between units.
• Risk identification: Identify potential risks when comparing predicted genetic values across disparate units due to low connectedness.

Methodology:

Implemented as an R package, GCA computes connectedness statistics from pedigree and genomic data using functions of prediction error variance (PEV) and the variance of unit effect estimates (VE).