Statistical Tool Identifies Genetic Changes Behind Neurological Conditions

Although scientists have long identified genes associated with these diseases, confirming which genetic alterations cause them has proven elusive. A key obstacle is the presence of unmeasured confounders — subtle, hidden biological or environmental factors, such as cell cycle stages or experimental conditions, that can affect cellular behavior and obscure true causality. Traditional CRISPR-based experiments, which modify genes and compare the results to unmodified cells, fail to account for these confounding variables. Now, researchers have developed a new statistical tool that could help pinpoint the genetic changes that cause diseases like Alzheimer’s and schizophrenia.

The statistical software tool, called causarray, was developed by researchers at Carnegie Mellon University (CMU, Pittsburgh, PA, USA), in collaboration with the University of Pittsburgh (Pittsburgh, PA, USA), after they became inspired by recent advances in causal inference. The tool utilizes the concept of counterfactual—a statistical method that estimates what would have happened to a treated cell had it not been treated. Causarray is among the first tools to apply this counterfactual framework to genomics. It also analyzes large-scale gene expression data to model what would occur in control cells, allowing it to identify common patterns and adjust for confounders across multiple genes. By doing so, it improves the precision of identifying genetic causes of disease from CRISPR and other genomic data.

Causarray has already demonstrated its effectiveness in pinpointing significant genetic changes linked to diseases. By correcting for unmeasured confounders and using advanced statistical modeling, the tool moves genetic research from association to causation. Though not explicitly stated, the study’s findings are poised to impact future genomic research published in peer-reviewed journals. Causarray is expected to play a vital role in unlocking causal mechanisms behind neurological conditions, ultimately informing therapeutic research and advancing personalized medicine. The developers plan to expand its integration into large-scale studies of brain disorders and other diseases.

"Recent advances, like CRISPR, hold the promise to lead to real breakthroughs in our understanding of brain disorders, but we will only achieve these advances if they are paired with powerful statistical tools. This is the magic of it," said Kathryn Roeder, co-author of the study.

Related Links:
CMU
University of Pittsburgh