Directional Association Measurement in Contingency Tables: Genomic Case

Area of Science:

• Bioinformatics
• Computational Biology
• Genomics

Background:

• Analyzing large datasets presents significant challenges in various scientific fields.
• Developing novel methods for data analysis is crucial for advancing research.
• Understanding associations within complex biological data, such as gene mutations and cancer types, requires robust analytical tools.

Purpose of the Study:

• To introduce and evaluate the Z association method for describing directional associations in contingency tables.
• To demonstrate the utility of the Z association method in analyzing large-scale biological data, specifically gene mutations across cancer types.
• To present the "MEUSassociation" R library as a tool for implementing the Z association method.

Main Methods:

• The Z association method was developed to analyze directional associations in contingency tables.
• The method allows for arbitrary grouping of categories for the two variables in a contingency table.
• The "MEUSassociation" R library was created to compute the standard Z coefficient and singular coefficients Z(A:B) for comprehensive analysis.

Main Results:

• Application of the Z coefficient to a dataset of gene mutations in different cancer types revealed significant associations.
• Results indicated relationships between specific gene mutations, annotation groups, and cancer types, aligning with existing cancer genomics knowledge.
• The ranked list of singular coefficients effectively reduced the complexity of the large-scale dataset.

Conclusions:

• The Z association method and its R implementation ("MEUSassociation") are valuable tools for categorical data analysis.
• This method offers a powerful approach to uncovering directional associations in complex, large-scale datasets.
• The findings support the utility of the Z association method in cancer genomics research for identifying key gene-cancer relationships.