G-DIRT

G-DIRT identifies duplicate germplasm using identity-by-state analysis of SNP genotyping data to support genebank curation and reduce redundancy that can bias genome-wide association studies.

Key Features:

• Identity-by-State Analysis: Uses identity-by-state analysis based on single nucleotide polymorphism (SNP) genotyping data to detect duplicate germplasms.
• Homozygous Genotypic Difference Threshold: Applies a homozygous genotypic difference threshold of 0.1% to define duplicates, with a reported accuracy of 94.97% on tetraploid wheat data.
• Dendrogram Visualization: Constructs dendrograms to represent genetic relationships among genotypes.
• Pre-processing Pipeline: Processes genotypic data through a comprehensive pre-processing pipeline to clean and standardize data prior to analysis.
• Offline R Interface: Provides an offline R interface implementation for analysis of high-dimensional genotypic data.

Scientific Applications:

• Genebank Management: Identification and removal of duplicate germplasms to focus conservation on unique genetic material.
• Genome-Wide Association Studies (GWAS): Prevention of biased GWAS results by reducing over-representation of duplicates in association panels.
• Research and Breeding Programs: Streamlining germplasm collections to prioritize genetically distinct samples for research and breeding.

Methodology:

Pre-processing of genotypic data followed by identity-by-state analysis of SNP genotyping data using a homozygous genotypic difference threshold of 0.1% (reported 94.97% accuracy in tetraploid wheat) and construction of dendrograms.