Technique for studying how proteins bind to DNA is often misused: Researchers propose a solution

Researchers at the University of California, San Diego, have published new guidelines that could help scientists dramatically improve their results when quantifying interactions between DNA and proteins. Understanding these interactions is critical to our understanding of human biology in general and can also help scientists develop new treatments for a wide range of diseases, including many cancers.

The article was published on September 13, 2024 in Biotechnology of nature

The researchers focused on peak normalization, a widely used molecular biology technique that helps ensure accurate and reliable results. Peak normalization involves adding a known amount of chromatin (DNA and associated proteins) to a sample before studying it, which helps researchers account for variations between multiple samples when compared to each other. Peak normalization is particularly useful for comparing two conditions, such as assessing the impact of a drug (by comparing a treated patient to an untreated patient) or of a mutation or deletion in a key gene (by comparing a normal patient to a mutant patient).

By exploring publicly available datasets that use peak normalization, the researchers were able to identify common scenarios in which peak normalization is easily misused and in which incorporating additional quality control measures and other “safeguards” into the technique could improve results or prevent misinterpretation.

By reanalyzing these data and conducting additional experiments, the team was able to come up with a list of nine key recommendations for researchers using peak normalization that could significantly increase the accuracy of their results. These measures include implementing consistent quality control measures, following best practices for computational analysis, and validating results with other analysis techniques.

“Many studies use peak normalization, and our results challenge the biological conclusions drawn from this approach,” said lead author Alon Goren, Ph.D. “Our recommendations can help address some of the pitfalls of peak normalization so that we can still reap the benefits of this valuable technique.”

The paper was authored by Lauren Patel and Yuwei Cao of UC San Diego and Eric Mendenhall, Ph.D., of the HudsonAlpha Institute for Biotechnology. The study was co-led by Alon Goren, Ph.D., and Christopher Benner, Ph.D., both associate professors in the Department of Medicine at the UC San Diego School of Medicine.