Asosiasi Single Nucleotide Polymorphism pada Diabetes Mellitus Tipe 2 Menggunakan Random Forest Regression

• Main Authors: Lina Herlina Tresnawati, Wisnu Ananta Kusuma, Sony Hartono Wijaya, Lailan Sahrina Hasibuan
• Format: Article
• Language: English
• Published: Universitas Gadjah Mada 2019-11-01
• Series: Jurnal Nasional Teknik Elektro dan Teknologi Informasi
• Subjects: diabetes mellitus tipe 2; epistatis; pemetaan asosiasi; random forest regression; single nucleotide polymorphism
• Online Access: http://ejnteti.jteti.ugm.ac.id/index.php/JNTETI/article/view/531

Precision medicine can be developed by determining association between genomic data, represented by Single Nucleotide Polymorphism (SNP), and phenotype of diabetes mellitus type 2 (T2D). The number of SNP is actually very abundance. Thus, sorting and filtering the SNP is required before conducting association. The purpose of this paper was to associate SNP with T2D phenotypes. SNP ranking was conducted to choose significant SNPs by calculating importance score. Selected SNPs were associated with T2D phenotype using random forest regression. Moreover, the epistasis was also examined to show the interactions among SNPs affecting phenotype. This paper obtained 301 importance SNPs. Top ten SNPs have association with five T2D protein candidates. The evaluation results of the proposed models showed the Mean Absolute Error (MAE) of 0.062. This results indicate the success of random forest regression in conducting SNP and phenotype association and epistatic examination between two SNPs