Credit: Florida Atlantic University
Cancers originating from different primary sample sites have specific patterns of miRNA expression, as revealed by principal component analysis, a popular technique for analyzing large datasets. These patterns allow for the highly accurate classification of cancer types by machine learning models.