Seon-Kyu Kim1, Jong Hwan Kim1, Seok-Joong Yun2, Wun-Jae Kim2, Seon-Young Kim1. 1. Medical Genomics Research Center, Korean Bioinformation Center, Korea Research Institute of Bioscience and Biotechnology, Department of Functional Genomics, University of Science and Technology, Daejeon 305-806, Korea and Department of Urology, Chungbuk National University College of Medicine, Cheongju 360-100, Korea Medical Genomics Research Center, Korean Bioinformation Center, Korea Research Institute of Bioscience and Biotechnology, Department of Functional Genomics, University of Science and Technology, Daejeon 305-806, Korea and Department of Urology, Chungbuk National University College of Medicine, Cheongju 360-100, Korea. 2. Medical Genomics Research Center, Korean Bioinformation Center, Korea Research Institute of Bioscience and Biotechnology, Department of Functional Genomics, University of Science and Technology, Daejeon 305-806, Korea and Department of Urology, Chungbuk National University College of Medicine, Cheongju 360-100, Korea.
Abstract
SUMMARY: Because cancer has heterogeneous clinical behaviors due to the progressive accumulation of multiple genetic and epigenetic alterations, the identification of robust molecular signatures for predicting cancer outcome is profoundly important. Here, we introduce the APPEX Web-based analysis platform as a versatile tool for identifying prognostic molecular signatures that predict cancer diversity. We incorporated most of statistical methods for survival analysis and implemented seven survival analysis workflows, including CoxSingle, CoxMulti, IntransSingle, IntransMulti, SuperPC, TimeRoc and multivariate. A total of 236 publicly available datasets were collected, processed and stored to support easy independent validation of prognostic signatures. Two case studies including disease recurrence and bladder cancer progression were described using different combinations of the seven workflows. AVAILABILITY AND IMPLEMENTATION: APPEX is freely available at http://www.appex.kr. CONTACT: kimsy@kribb.re.kr SUPPLEMENTARY INFORMATION: Supplementary data are available at Bioinformatics online.
SUMMARY: Because cancer has heterogeneous clinical behaviors due to the progressive accumulation of multiple genetic and epigenetic alterations, the identification of robust molecular signatures for predicting cancer outcome is profoundly important. Here, we introduce the APPEX Web-based analysis platform as a versatile tool for identifying prognostic molecular signatures that predict cancer diversity. We incorporated most of statistical methods for survival analysis and implemented seven survival analysis workflows, including CoxSingle, CoxMulti, IntransSingle, IntransMulti, SuperPC, TimeRoc and multivariate. A total of 236 publicly available datasets were collected, processed and stored to support easy independent validation of prognostic signatures. Two case studies including disease recurrence and bladder cancer progression were described using different combinations of the seven workflows. AVAILABILITY AND IMPLEMENTATION: APPEX is freely available at http://www.appex.kr. CONTACT: kimsy@kribb.re.kr SUPPLEMENTARY INFORMATION: Supplementary data are available at Bioinformatics online.