Open AccessA comparative study of different classification algorithms on RNA-Seq cancer data
Author(s) -
Nihat Yılmaz ŞİMŞEK,
Bülent Haznedar,
Cihan Kuzudisli
Publication year - 2020
Publication title -
new trends and issues proceedings on advances in pure and applied sciences
Language(s) - English
Resource type - Journals
ISSN - 2547-880X
DOI - 10.18844/gjpaas.v0i12.4983
Subject(s) - rna seq , cancer , kidney cancer , gene , random forest , rna , computational biology , computer science , gene expression , algorithm , bioinformatics , artificial intelligence , biology , genetics , transcriptome
Gene mutations are the most important reason of cancer diseases, and there are different kind of causing genes across these diseases. RNA-Seq technology enables us to allow for gathering information about many genes simultaneously; hence, RNA-Seq data can be used for cancer diagnosis and classification. In this study, RNA-Seq dataset for renal cell cancer is analysed using three different developed classification methods: random forest (RF), artificial neural network (ANN) and deep learning (DL). The genes in our dataset are related to the following cancer types: kidney renal papillary cell, kidney renal clear cell and kidney chromophore carcinomas. It suggests that the DL method gives the highest accuracy rate compared to RF and ANN for 95.15%, 91.83% and 89.22%, respectively. We believe that the results acquired in this study will make a contribution to the classification of cancer types and support doctors in their processes of decision making.
Keywords: Classification, gene-expression, RNA-Seq, DL.