|
Peter Ulz 1,*, Gerhard Thallinger 2 , Martina Auer 1 , Ricarda Graf 1 , Karl Kashofer 3 , Stephan Wenzel Jahn 3, Luca Abete3, Gunda Pristauz4, Edgar Petru4, Jochen Geigl1, Ellen Heitzer1, Michael Speicher1 1 Institute of Human Genetics, Medical University Graz, Austria 2 Institute of Molecular Biotechnology, University of Technology, Graz, Austria 3 Institute of Pathology, Medical University of Graz, Graz, Austria 4 Department of Obstetrics and Gynecology, Medical University of Graz, Graz, Austria |
Abstract
Cell-free DNA (cfDNA) consists predominantly of nucleosome- protected DNA shed into the bloodstream by cells undergoing apoptosis. By whole-genome sequencing of cfDNA fragments two regions which inform about nucleosome occupancy were identi- fied. Since nucleosome occupancy is a marker of gene expression at transcription start sites (TSS), read depth differences may also inform about gene expression.
By a machine learning approach, gene expression was predicted from a pool of non-cancer controls and compared to previous gene expression results from studies of circulating RNA in healthy donors.
In tumor patients, this approach was applied to circulat- ing tumor DNA (ctDNA) whole-genome sequencing from regions exhibiting copy-number alterations. Gene expression prediction was compared to RNA-Seq from matching primary tumors and yielded high concordance.
Our analyses provide functional information about tumor cells from DNA sequencing and adds an additional layer of information to the analysis of ctDNA.
http://dx.doi.org/10.1016/j.bdq.2017.02.028
| Back to Liquid Biopsies and Molecular diagnostic 2 |
|---|
