Reliability of statistical associations between genes and disease

Area of Science:

• Genetics
• Biostatistics
• Genomic Epidemiology

Background:

• Many genetic associations with diseases lack reproducibility.
• This issue stems from insufficient replication and poor understanding of statistical significance's link to reproducibility.

Purpose of the Study:

• To introduce a novel classification system, the Better Associations for Disease and GEnes (BADGE) system.
• To provide a framework for evaluating the reproducibility of genetic association studies.

Main Methods:

• Proposes a five-class system (BADGE classes) for genetic associations.
• Classification is based on the P value threshold of the statistical association.

Main Results:

• A first-class association (P < 2 x 10(-7)) suggests high reproducibility.
• A fifth-class association (P < 5 x 10(-2)), conventional significance, offers minimal reproducibility assurance.
• Intermediate classes (second, third, fourth) are defined by P value ranges.

Conclusions:

• The BADGE system offers a standardized method to assess genetic association reproducibility.
• This classification aids researchers in prioritizing and interpreting genetic findings for disease association studies.