Diseases like cancer and neurodegenerative disorders often start with genetic mistakes, but turning that knowledge into effective treatments has historically been like trying to fix a car when you've identified 300 different faulty parts, all doing different things. A new study published in *Nature* introduces a potential solution: a platform called PerturbFate, which can systematically track how disease-related genetic changes alter cells and identify where those changes ultimately converge.
"We focus here on cancer drug resistance, but the paper really starts from a broader question: once you know that a disease is associated with hundreds of genes, how do you design one therapy to target it?" says Junyue Cao, head of the Laboratory of Single-Cell Genomics and Population Dynamics. "We wondered whether all these different genes may be mediated by some shared downstream signaling that we can discover and target instead."
Advances in genetic sequencing have allowed scientists to identify large numbers of disease-linked mutations, but this progress created a major new challenge: the genes involved often perform very different jobs inside cells, including controlling gene activity and managing cell signaling pathways. Because of this complexity, designing treatments that address many mutations at once has been difficult. Cao suspected that these seemingly unrelated mutations might not actually act independently - they could funnel into shared downstream programs that ultimately determine how cells behave. If that were true, scientists would not need to target every mutation separately; they could focus on common regulatory nodes that control the disease process.
To test that, the team needed a system capable of comparing many genetic disruptions at the same time while monitoring how each one reshaped a cell in detail. Existing technologies could only capture part of the picture, often measuring one layer of cellular activity at a time or missing how gene activity changes dynamically over time. Graduate student Zihan Xu developed PerturbFate to overcome those limitations, enabling researchers to observe how different genetic disruptions alter cells in real time by simultaneously tracking DNA accessibility and RNA production. Because these measurements are collected within the same single cell, the system can reveal the gene networks controlling cell behavior and identify where distinct mutations produce the same downstream effects.
To test the platform, the researchers turned to melanoma, where many different mutations can produce resistance to treatment. The team selected 143 genes previously associated with resistance to the melanoma drug Vemurafenib and systematically disabled them in melanoma cells. PerturbFate then monitored how each disruption changed cellular behavior over time. By labeling newly produced RNA, the researchers could separate fresh gene activity from older molecular signals. Single-cell profiling also allowed them to track which genes were active, which regions of DNA became accessible, and how those changes evolved.
After examining more than 300,000 cells, the researchers found that many different mutations consistently pushed melanoma cells into the same drug-resistant condition. When the team targeted the shared regulatory control points driving that state, drug resistance dropped significantly, suggesting a promising strategy for combination therapies. The study also uncovered an important detail involving the Mediator Complex, a cellular structure that helps regulate gene activity. Researchers found that disrupting different parts of this same complex could trigger drug resistance through entirely different biological routes, but those pathways still converged on the same melanoma survival signal known as VEGFC. When researchers blocked VEGFC, the resistant melanoma cells were no longer able to grow.
"We're capturing not just gene expression, but also RNA dynamics and chromatin state," says Cao. "That's critical for identifying the upstream regulators that drive these disease states." The findings suggest that even highly complex genetic diseases may rely on shared vulnerabilities that can be therapeutically targeted. Rather than designing separate treatments for every mutation, scientists may be able to focus on common regulatory pathways that multiple mutations depend on. The researchers have made both the laboratory and computational tools behind PerturbFate publicly available, and they now plan to expand the approach beyond cultured cells and apply it to living systems. Cao and his colleagues hope to use the technology to study conditions such as aging and Alzheimer's disease, both major research areas in his lab.
"This is just a starting point," says Cao. "Now that we've demonstrated the approach in a simple model, we're working to extend it into living systems to study even more complex diseases."