View Item 
      •   IPB Repository
      • Dissertations and Theses
      • Undergraduate Theses
      • UT - Faculty of Mathematics and Natural Sciences
      • UT - Computer Science
      • View Item
      •   IPB Repository
      • Dissertations and Theses
      • Undergraduate Theses
      • UT - Faculty of Mathematics and Natural Sciences
      • UT - Computer Science
      • View Item
      JavaScript is disabled for your browser. Some features of this site may not work without it.

      Asosiasi Single Nucleotide Polymorphism dan Fenotipe pada Penyakit Diabetes Mellitus Tipe 2 Menggunakan Metode Elastic Net

      Thumbnail
      View/Open
      Cover (800.3Kb)
      Fullteks (1.630Mb)
      Lampiran (443.9Kb)
      Date
      2022
      Author
      Romdendine, Muhammad Fahrury
      Kusuma, Wisnu Ananta
      Metadata
      Show full item record
      Abstract
      Precision Medicine bertujuan untuk mendesain strategi pengobatan maupun pencegahan paling tepat terhadap suatu individu berdasarkan profil genetiknya. Profil genetik bisa didapatkan melalui penanda genetik yang bisa diidentifikasi menggunakan metode machine learning. Data penanda genetik seperti data Single Nucleotide Polymorphism (SNP) memiliki permasalahan high dimensionality data yang tidak bisa diatasi oleh metode regresi linear biasa. Pada penelitian ini digunakan metode regularisasi yaitu Elastic Net untuk membangun model prediktif untuk memprediksi nilai insulin tolerance yang merupakan fenotipe Type 2 Diabetes Mellitus (T2DM) sekaligus menyeleksi SNP yang signifikan. Hasil yang didapat menunjukkan Elastic Net mampu menyeleksi 201 SNP signifikan dari 690 SNP dengan galat yang rendah (MAE = 0,619) dan koefisien determinasi yang baik (R2 = 0,99). Sebanyak 201 SNP yang terseleksi merepresentasikan 36 gen yang paling berasosiasi dari 68 gen yang ada pada data. Sebanyak 77,8% gen yang didapatkan berhasil divalidasi menggunakan studi literatur sehingga dapat dikatakan Elastic Net memiliki kemampuan yang baik dalam mengasosiasikan SNP-Fenotipe T2DM. Hasil penelitian ini bermanfaat untuk membantu mengurangi ruang pencarian kandidat penanda genetik T2DM.
       
      Precision medicine aimed to design the most precise strategy to medicate or prevent someone from a disease based on an individual genetic profile. Genetic profile could be obtained through machine learning identification of genetic marker. Genetic data such as single nucleotide polymorphism (SNP) has a high dimensionality problem that traditional linear regression could not overcome. Regularized method named Elastic Net were utilized in this research to build predictive model of insulin tolerance value which is a type 2 diabetes mellitus’s (T2DM) phenotype marker and selects most significant SNPs simultaneously. The results shown that Elastic Net selected 201 SNPs out of 690 SNPs with low error (MAE = 0.619) and good coefficient of determination (R2 = 0,99). 201 selected SNPs represented 36 most associated genes out of 68 genes. 77.8% genes were successfully validated using literature research thus proofing the performance of Elastic Net. The results are useful to narrow the search space of T2DM’s genetic marker candidates thus wet lab experiment would become more efficient.
       
      URI
      http://repository.ipb.ac.id/handle/123456789/112860
      Collections
      • UT - Computer Science [2482]

      Copyright © 2020 Library of IPB University
      All rights reserved
      Contact Us | Send Feedback
      Indonesia DSpace Group 
      IPB University Scientific Repository
      UIN Syarif Hidayatullah Institutional Repository
      Universitas Jember Digital Repository
        

       

      Browse

      All of IPB RepositoryCollectionsBy Issue DateAuthorsTitlesSubjectsThis CollectionBy Issue DateAuthorsTitlesSubjects

      My Account

      Login

      Application

      google store

      Copyright © 2020 Library of IPB University
      All rights reserved
      Contact Us | Send Feedback
      Indonesia DSpace Group 
      IPB University Scientific Repository
      UIN Syarif Hidayatullah Institutional Repository
      Universitas Jember Digital Repository