Related Experiment Video
Updated: Sep 10, 2025
11:02
Detecting Somatic Genetic Alterations in Tumor Specimens by Exon Capture and Massively Parallel Sequencing
Published on: October 18, 2013
19.5K
Leveraging sequences missing from the human genome to diagnose cancer
Ilias Georgakopoulos-Soares1,2,3, Ofer Yizhar-Barnea4,5, Ioannis Mouratidis6,7
1Department of Bioengineering and Therapeutic Sciences, University of California San Francisco, San Francisco, CA, USA. izg5139@psu.edu.
Communications Medicine
|August 21, 2025
Summary
A new prediction model using neomers, short DNA sequences from tumor mutations, accurately detects various cancers, including early stages, using cell-free DNA. This tool identifies cancer subtypes and regulatory mutations, improving diagnostic sensitivity and specificity.
Area of Science:
- Genomics
- Molecular Biology
- Cancer Research
Background:
- Cell-free DNA (cfDNA) analysis offers promise for cancer diagnosis, treatment, and survival.
- Existing cfDNA diagnostic methods face technical limitations impacting accuracy and scope.
Purpose of the Study:
- To develop a novel prediction model for cancer detection using cfDNA.
- To identify specific DNA signatures, termed neomers, associated with tumor-associated mutations.
Main Methods:
- Developed a prediction model utilizing neomers, defined as 13-17 nucleotide DNA sequences.
- Neomers are predominantly absent in healthy individuals and arise from tumor-specific mutations.
- Analyzed 2577 cancer genomes across 21 types and 465 cfDNA whole-genome sequences.
Main Results:
- Neomer-based classifiers accurately detected cancer, including early stages, and distinguished subtypes.
- Neomers demonstrated higher accuracy in distinguishing tumor types compared to state-of-the-art methods.
- Precisely detected lung and ovarian cancer from cfDNA with AUCs of 0.89-0.94.
- Identified cancer-associated mutations impacting gene regulatory activity through reporter assays.
Conclusions:
- Identified neomers as a sensitive, specific, and simple tool for cancer diagnostics.
- The neomer approach can detect cancer-associated mutations within gene regulatory elements.
- This method enhances the potential of cfDNA for non-invasive cancer detection and characterization.
