Johns Hopkins researchers and collaborators demonstrated a prototype liquid biopsy that uses genome‑wide cell‑free DNA (cfDNA) fragmentation patterns and AI to detect liver fibrosis and cirrhosis. Published in Science Translational Medicine, the study showed fragmentomic signatures in plasma that correlate with organ‑specific injury and chronic liver disease states. The team, led by Victor Velculescu and partners including Delfi Diagnostics, trained models to identify fragment‑pattern changes tied to liver pathology and immune‑related shifts in white blood cell composition. Results suggest cfDNA fragmentomics can extend liquid biopsy beyond oncology to chronic organ diseases and could offer a noninvasive alternative to liver biopsy for staging fibrosis. Fragmentomics analyzes how cfDNA fragmentation and genome‑wide coverage reflect chromatin structure and cell‑type of origin; coupling that signal with machine learning can reveal tissue‑specific injury patterns without targeting individual mutations, broadening clinical applications for cfDNA assays.