Bilkent University
Department of Computer Engineering


PAMOGK-Web: A Framework for Cancer Sub-Type Identification using Copy Number Variations


Furkan Mustafa Akdemir
MS Student
(Supervisor:Asst. Prof. Dr. A. Ercüment Çiçek)
(Co-Supervisor: Asst. Prof. Dr. Öznur Taştan Okan (Sabancı Univ.))

Computer Engineering Department
Bilkent University

Detection of different sub-groups of patients in diseases caused by genetic causes is important for development of therapeutics and also the underlying causes. Cancer studies throughout years collected a large dataset composed of different omics data such as genome, transcriptome, metabolome, proteome, etc. which are referred as multi-omics data. The multi-omics data provides multiple views into the cancer which can be used to find underlying causes from different perspectives and capture relations not possible with a single view approach. Previously we developed a pipeline that uses multi-omics data to detect sub-groups of patients called PAMOGK. PAMOGK used gene pathways as a foundation and multi-omics data as information to generate graph kernels and using multi-view clustering grouped patients into informative sub-groups. PAMOGK was designed as a general framework that can be used to map many different omics data and in the first part of this work, PAMOGK was extended with copy number variation data with aim to improve the accuracy of sub-groups.


DATE: 14 December 2020, Monday @ 17:40