New Gene Discovery Methods Enhance Understanding of Disease Biology

A recent study has revealed that two primary approaches for discovering disease-related genes offer distinct insights into biological processes. This research, published in the journal Nature, underscores the implications these methods hold for drug development and understanding genetic influences on health.

The investigation was led by a collaborative team from NYU Langone Health, Stanford University, UC San Francisco, and the University of Tokyo. It examined the human genome, which contains thousands of genes responsible for producing proteins and regulatory DNA that dictates gene activity. The study focused on the impact of small DNA differences, known as variants, on various traits like height, hair color, and disease susceptibility.

Two gene discovery methods were analyzed: genome-wide association studies (GWAS) and burden tests. GWAS examines common variants throughout the genome to identify those linked to diseases, while burden tests concentrate on rare variants that alter proteins. By reviewing the results from both methods across 209 traits derived from the UK Biobank, which includes genetic data from hundreds of thousands of individuals, researchers found significant differences in the types of genes identified.

Hakhamanesh Mostafavi, Ph.D., an assistant professor in the Department of Population Health at NYU Grossman School of Medicine and co-senior author of the study, noted that burden tests tend to reveal genes that predominantly impact the specific disease under investigation, with minimal influence on other traits. Conversely, GWAS identifies both disease-specific genes and those that may affect multiple diseases and biological processes.

The findings highlight a critical distinction in the results produced by these methods. The study suggests that some genes specifically influence a single trait, while others have broader implications. Variants that significantly disrupt “multi-trait” genes tend to be less common due to evolutionary pressures that favor survival and reproduction. This rarity complicates their detection in burden tests. In contrast, GWAS can identify these genes since regulatory DNA variants may affect gene function in less dramatic ways, allowing them to persist through generations.

The researchers propose that two key features should guide gene prioritization related to disease risks: “importance,” which measures the extent of a gene’s impact on disease, and “specificity,” which indicates whether a gene primarily affects one condition or multiple traits. Understanding these factors could aid in pinpointing the most promising therapeutic targets and anticipating potential side effects.

A noteworthy aspect of the study relates to the significance of p-values, a statistical measure used to determine the reliability of study results. The research reveals that the p-values derived from GWAS and burden tests do not reliably indicate a gene’s importance. This insight is pivotal, as accurately identifying vital genes can illuminate the biological processes central to disease mechanisms.

Mostafavi emphasized that while GWAS and burden tests yield valuable insights regarding gene importance, they have not been interpreted in this manner previously. He advocates for the development of new methods that could enhance the prioritization of genes based on their biological characteristics.

Looking ahead, the research team has commenced work on innovative methods to assess gene importance. Currently, neither GWAS nor burden tests possess adequate power to definitively gauge the influence of individual genes on disease. However, by integrating these findings with an expanding body of experimental data detailing gene functions at the cellular level, the authors suggest that machine learning techniques could uncover shared patterns among genes, thereby refining estimates of their significance.

As co-senior author Jeffrey Spence, Ph.D., an assistant professor at UC San Francisco, noted, this approach could revolutionize the field by enabling researchers to utilize comprehensive cell-level data to better understand human traits, identify critical disease-related genes, and streamline drug development processes.

The study signifies a pivotal step forward in gene discovery methods, reinforcing the idea that a nuanced understanding of genetic factors is essential for advancing medical research and therapeutic strategies.