CancerLocator: non-invasive cancer diagnosis and tissue-of-origin prediction using methylation profiles of cell-free DNA

• Published in: Genome Biology Vol. 18; no. 1; p. 53
• Main Authors: Kang, Shuli, Li, Qingjiao, Chen, Quan, Zhou, Yonggang, Park, Stacy, Lee, Gina, Grimes, Brandon, Krysan, Kostyantyn, Yu, Min, Wang, Wei, Alber, Frank, Sun, Fengzhu, Dubinett, Steven M., Li, Wenyuan, Zhou, Xianghong Jasmine
• Format: Journal Article
• Language: English
• Published: London BioMed Central 24.03.2017; Springer Nature B.V
• Subjects: Algorithms; Animal Genetics and Genomics; Bioinformatics; Biomarkers; Biomedical and Life Sciences; blood sampling; Breast cancer; Cancer; Computer Simulation; CpG Islands; Deoxyribonucleic acid; DNA; DNA Methylation; DNA sequencing; DNA, Neoplasm - blood; DNA, Neoplasm - genetics; Epigenesis, Genetic; Epigenomics - methods; Evolutionary Biology; Gene expression; Genomes; Human Genetics; Humans; Life Sciences; Medical diagnosis; Medical screening; Method; Methods; Microbial Genetics and Genomics; Models, Statistical; neoplasms; Neoplasms - diagnosis; Neoplasms - genetics; Pancreatic cancer; Patients; Plant Genetics and Genomics; prediction; Reproducibility of Results; Tumors; Workflow; Cell-free DNA; DNA methylation; Liquid biopsy; Next-generation sequencing; Cancer diagnosis
• ISSN: 1474-760X; 1474-7596; 1474-760X
• DOI: 10.1186/s13059-017-1191-5

We propose a probabilistic method, CancerLocator, which exploits the diagnostic potential of cell-free DNA by determining not only the presence but also the location of tumors. CancerLocator simultaneously infers the proportions and the tissue-of-origin of tumor-derived cell-free DNA in a blood sample using genome-wide DNA methylation data. CancerLocator outperforms two established multi-class classification methods on simulations and real data, even with the low proportion of tumor-derived DNA in the cell-free DNA scenarios. CancerLocator also achieves promising results on patient plasma samples with low DNA methylation sequencing coverage.