Application of text mining and data mining in cancer research

Abstract

This study aimed to use text and data mining to select candidate genes which are associated with liver and colorectal cancer Fifty unknown candidate genes were selected by data mining the EST library and ZNF496 RMI2 and U41 were found that may be associated with WNT target genes and liver cancer Twenty known candidate genes were selected by text mining PubstractHelper and StemTextSearch and data mining GeneCards and GEO of NCBI IGF2BP1 was found to be associated with the oncofetal circulating cancer stem cell-like markers associated with the recurrence of hepatocellular carcinoma by experiment Twenty-three candidate genes were selected by data mining from TCGA (the cancer genome atlas) data and the three remaining candidate genes are examined as to whether they are expressed in colorectal cancer by IHC (immunohistochemistry) This study shows that text and data mining are alternative methods to help scientist narrow down their candidate genes which are associated with cancer

Date of Award	2017 Jul 10
Original language	English
Supervisor	Chung-Liang Ho (Supervisor)