Min-Seok Kwon1, Yongkang Kim2, Seungyeoun Lee3, Junghyun Namkung4, Taegyun Yun4, Sung Gon Yi4, Sangjo Han4, Meejoo Kang5, Sun Whe Kim5, Jin-Young Jang6, Taesung Park7,8. 1. Interdisciplinary program in Bioinformatics, Seoul National University, Seoul, Korea. 2. Department of Statistics, Seoul National University, Seoul, Korea. 3. Department of Mathematics and Statistics, Sejong University, Seoul, Korea. 4. Immunodiagnostics R&D Team, IVD Business Unit, New Business Division, SK telecom Co, Seongnam, Korea. 5. Department of Surgery, Seoul National University Hospital, Seoul, Korea. 6. Department of Surgery, Seoul National University Hospital, Seoul, Korea. jangjy4@gmail.com. 7. Interdisciplinary program in Bioinformatics, Seoul National University, Seoul, Korea. tspark@stats.snu.ac.kr. 8. Department of Statistics, Seoul National University, Seoul, Korea. tspark@stats.snu.ac.kr.
After publication of this article [1] it was brought to the Editors’ attention that there were inaccurate descriptions of validation data in its abstract and list of abbreviations. For the validation, we used the three datasets from Gene Expression Omnibus (GEO), not the Cancer Genome Atlas (TCGA).The corrected description for line 8–10 in abstract is as below:“For selecting even more reliable and robust markers, we performed validation by independent datasets from the Gene Expression Omnibus (GEO) data depository.”In page 9, the corrected description of list of abbreviations used is as follows:“AUC, Area under curve; BA, Balanced accuracy; BR, Breast cancer; GEO, Gene Expression Omnibus; GO, Gene ontology; HCC, Hepatocellular carcinoma; LC, Lung cancer; LOOCV, Leave-one-out cross-validation; LP, Lymphoma; mRNA, messenger RNA; miRNA, microRNA; PDAC, Pancreatic ductal adenocarcinoma; SVM, Support vector machine;”