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.

      Prediksi Kandidat Senyawa Herbal Sebagai Obat Inflamasi COVID-19 Menggunakan Pemodelan Deep Semi-Supervised Learning

      Thumbnail
      View/Open
      Cover (884.2Kb)
      Lampiran (1.740Mb)
      Fullteks (1.459Mb)
      Date
      2021
      Author
      Khalid, Irfan Alghani
      Kusuma, Wisnu Ananta
      Priandana, Karlisa
      Metadata
      Show full item record
      Abstract
      COVID-19 merupakan sebuah penyakit yang disebabkan oleh virus SARS-CoV-2. Peradangan akut (hiperinflamasi) merupakan salah satu gejala dari COVID-19. Gejala tersebut dapat memperparah kondisi pasien COVID-19. Penelitian ini membangun model Stacked Autoencoders-Deep Neural Network (SAE-DNN) yang nantinya digunakan untuk memprediksi kandidat senyawa herbal sebagai obat anti inflamasi COVID-19. Selain itu, dilakukan perbandingan performa model berdasarkan representasi data yang berbeda. Alur penelitian dimulai dari pengambilan data, praproses data, pemodelan, dan diakhiri dengan pengujian model pada data uji untuk mendapatkan kandidat senyawa herbal. Penelitian ini menunjukkan bahwa pemodelan menggunakan SAE-DNN dengan representasi senyawa menggunakan fingerprint dan protein menggunakan Dipeptide Composition (DC) menghasilkan performa terbaik dengan akurasi sebesar 0,98857, precision sebesar 0,96722, recall sebesar 0,96419, AUROC sebesar 0,99596, dan F1 Score sebesar 0,96567. Dengan pemodelan tersebut, didapatkan senyawa herbal yang berinteraksi dengan protein target sebanyak 33 senyawa.
       
      COVID-19 is a disease caused by the SARS-CoV-2 virus. Hyper inflammation is one of the COVID-19 symptoms that worsen a person’s condition. This research aims to build a Stacked Autoencoders-Deep Neural Network (SAE-DNN) model, which will be used for finding herbal drug candidates as a COVID-19 anti-inflammatory drug. The model performance is compared based on different data representations. The research flow starts from data collection, data preprocessing, modelling, and finally test the model on the herbal data to obtain candidate herbal compounds. This study shows that modelling using SAE-DNN with the representation of compounds using fingerprints and proteins using Dipeptide Composition (DC) produces the best performance with an accuracy of 0,96722, a recall of 0,96419, AUROC of 0,99596, and an F1 score of 0,96567. From this modelling, around 33 herbal compounds are found as candidate drug for anti-inflammation.
       
      URI
      http://repository.ipb.ac.id/handle/123456789/109500
      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