The distribution-based p-value for the outlier sum in differential gene expression analysis

Area of Science:

• Genomics
• Statistical Genetics
• Bioinformatics

Background:

• Outlier sums were previously proposed for outlier gene detection.
• Existing methods lacked formal statistical inference and distributional properties.

Purpose of the Study:

• To propose a novel outlier sum for enhanced outlier gene detection.
• To develop the asymptotic distribution theory and formulate a p-value for the new method.
• To assess the efficiency of the proposed method compared to existing techniques.

Main Methods:

• Development of a new outlier sum statistic.
• Derivation of asymptotic distribution theory and analytic p-value formulation.
• Power comparison with existing outlier sum methods using large-sample theory.

Main Results:

• The proposed outlier sum method demonstrates superior efficiency in detecting outlier genes.
• The method was successfully applied to DNA microarray data from breast tumor samples.
• Formal statistical inference and distributional properties were established for the new outlier sum.

Conclusions:

• The novel outlier sum provides a statistically robust and efficient approach for outlier gene detection.
• This method enhances the analysis of genomic data, particularly in identifying disease-related gene expression anomalies.
• The findings suggest improved diagnostic and research capabilities in genomics.