GIPS

GIPS identifies genes associated with specific phenotypes by analyzing genomic sequences from multiple unrelated mutants within an integrated probabilistic framework for sequencing-based forward genetics.

Key Features:

• Integrated probabilistic framework: Uses a probabilistic model to integrate evidence and prioritize candidate genes from sequencing data of multiple mutants.
• Sequencing-based forward genetics: Enables cloning of phenotype-associated genes directly from mutant sequencing without requiring segregation populations.
• Analysis of multiple unrelated mutants: Compares genomic sequences from several unrelated mutants exhibiting the same phenotype to identify shared causal genes.
• Probability of reporting the true phenotype-associated gene: Estimates the likelihood that a reported candidate gene is genuinely associated with the phenotype.
• Expected number of random genes reported: Calculates the expected count of genes that might be reported by chance to help distinguish true positives from random hits.
• Significance of each candidate gene's association with the phenotype: Quantifies the statistical significance of the association between each candidate gene and the phenotype.
• Significance of violating Mendelian assumptions: Assesses implications when no gene is reported or when candidate genes fail validation, indicating potential deviations from expected inheritance patterns.
• Support for EMS mutants: Applicable to analysis of ethyl methanesulfonate (EMS) mutants, as demonstrated in empirical studies.

Scientific Applications:

• Cloning phenotype-associated genes: Identification of genes responsible for mutant phenotypes in sequencing-based forward genetics studies.
• Detection of epistatic and suppressor genes: Identification of genes that epistatically suppress phenotypes, including the reported example in rice (Oryza sativa) involving the phosphate2 mutant.
• Analysis of EMS-mutagenized populations: Application to discovery of causal variants in collections of EMS-induced mutants.

Methodology:

Analyzes genomic sequences from multiple unrelated mutants within an integrated probabilistic framework and estimates four measures: probability of reporting the true phenotype-associated gene, expected number of random genes reported, significance of each candidate's association with the phenotype, and significance of violating Mendelian assumptions.