Scan statistics to scan markers for susceptibility genes

Area of Science:

• Genetics
• Statistical genetics
• Genomic analysis

Background:

• Genome-wide association studies (GWAS) are crucial for identifying genetic loci associated with diseases.
• Conventional methods often analyze single genetic markers, potentially missing complex associations across linked markers.
• Combining information from contiguous genetic markers can enhance the power to detect susceptibility loci.

Purpose of the Study:

• To develop and validate a novel scan statistic method for combining information from multiple contiguous genetic markers in genome screens.
• To identify susceptibility loci for complex diseases using a dichotomous outcome variable.
• To assess the significance of identified regions using Monte Carlo permutation tests.

Main Methods:

• Application of scan statistics to aggregate data from contiguous genetic markers (e.g., allele sharing, lod scores).
• Focus on dichotomous outcome variables (case/control, affected/unaffected siblings).
• Significance assessment using Monte Carlo permutation tests to determine P values for varying scan statistic lengths.

Main Results:

• The developed scan statistic method identified a significant susceptibility region in a genome screen of autism families (genome-wide significance P = 0.038).
• This region was not detected using conventional single-marker analysis approaches.
• The method proved informative and yielded surprising results.

Conclusions:

• The novel scan statistic method offers a powerful approach for detecting genetic susceptibility loci by leveraging information from contiguous markers.
• This method can identify regions missed by traditional genome screening techniques.
• The findings highlight the potential of integrated marker analysis for complex disease genetics, as demonstrated in autism research.