PhD Scientific Days 2018

Budapest, April 19–20, 2018

ROC plotter: An Online Biomarker Validation Tool using Cancer Gene Expression Data and ROC analysis

Fekete, János

János Fekete1 , Balázs Győrffy1,2

1. Semmelweis University 2nd Department of Pediatrics, Budapest, Hungary.
2. MTA TTK Lendület Cancer Biomarker Research Group, Institute of Enzymology, Budapest, Hungary.

Language of the presentation

Hungarian

Text of the abstract

Introduction: Discovery of predictive biomarkers is an important issue in cancer research. In the last decades, several gene expression based tools were developed for breast cancer, however the primary focus of these tool was limited to assess the prognostic value of predictors.
Aims: Our goal was to develop an online tool to draw Receiver Operating Characteristics (ROC) plots, which can be used to evaluate the relevance of the expression levels of selected genes on treatment response in breast cancer.
Method: The background database of ROC plot is a manually curated. Gene expression data, treatment and response related information were downloaded from GEO and Array Express (Affymetrix HG-U133A, HG-U133A 2.0, and HG-U133 Plus 2.0 microarrays). Database is handled by a PostgreSQL server, which integrates gene expression and clinical data simultaneously. ROC plotter recognizes 54,675 Affymetrix probe set IDs and 70,632 gene symbols. From the MAS5 normalized microarray data ROC and box-whisker plots are generated. Significance is computed using a Mann-Whitney test.
Results: All together 3,104 samples were integrated into the system. The web-tool enables to run the analysis in real time. Patient selection includes filters for receptors status, lymph node involvement, histological grade, and molecular subtypes. The complete analysis tool can be accessed online at: www.rocplot.org.
Conclusion: In summary, we developed an online biomarker discovery tool that enables to assess the predictive power of 54,675 genes in 3,104 breast cancer patients.

Data of the presenter

Doctoral School: Pathological sciences
Program: Oncology
Supervisor: Balázs Győrffy
E-mail address: fekete.janos@med.semmelweis-univ.hu