Identification of Single Nucleotide Polymorphism using Support Vector Machine on Imbalanced Data

Abstract

Abstract-The advance of DNA sequencing tech110/ogy presents a sig11iflca11t bioi11for111atic chal/e11ges in a do11111strea111 analysis such as identification of single nucleotide poly111orphis111 (SNP), SNP is the most ab1111da11t form of genetic 111arker and hai•e been one of the 1110s/ crucial researches in bioinfor111alics. SNP has been applied in wide area, but analysis of SNP in plants is very• limited, as in cu//ivated soybean (Glycine lltll~ l.). This paper discusse; the identijicatio11 of SNP i11 culfil'aled soybean u.-;ing Support Vector Machine (SVM). SVM is trained using positfre and negative SNP. Previous/)1, we perfor111ed a balancing positive and negative SNP with u11dersa111pling and oversa111pling to obtain training <lata. As a result, the 111otlel 1vhich is trainetl 1vitfl balancetl tlatll has better perfor111a11ce tlta11 that with i111bala11ced tlllta.