New Tool to Improve Cell-Free DNA Testing Could Improve Disease Diagnosis Accuracy

A research team at UCLA Health (Los Angeles, CA, USA) has developed a comprehensive and high-resolution methylation atlas based on a huge dataset composed of 521 noncancerous tissue samples representing 29 major human tissue types. The researchers demonstrated that their method, termed cfSort, successfully identified specific methylation patterns unique to each tissue at the fragment level, and also validated these results using additional datasets. Taking their research a step further, the team highlighted the clinical applications of cfSort in two potential areas: assisting in disease diagnosis and tracking treatment side effects. By using cfSort to estimate the tissue-derived cfDNA fraction, the researchers could evaluate and predict clinical outcomes in patients.

“We have shown that the cfSort outperformed the existing methods in terms of accuracy and detection limit: making more accurate tissue fraction estimation and distinguishing a lower level of tissue-derived cfDNA,” said researcher Shuo Li. “In addition, the cfSort demonstrated nearly perfect robustness toward the unseen local fluctuations of tissue compositions, indicating its wide applicability to diverse individuals.”

Related Links:
UCLA Health