Prediksi Struktur Sekunder Protein Virus Influenza A Berbasis Machine Learning Menggunakan Pengkodean Data Asam Amino Dengan VHSE dan Sliding Window

Fanani, Sanada Aulia (2024) Prediksi Struktur Sekunder Protein Virus Influenza A Berbasis Machine Learning Menggunakan Pengkodean Data Asam Amino Dengan VHSE dan Sliding Window. Masters thesis, Institut Teknologi Sepuluh Nopember.

[thumbnail of 6002201018-Master_Thesis.pdf] Text
6002201018-Master_Thesis.pdf - Accepted Version
Restricted to Repository staff only until 1 October 2026.

Download (4MB) | Request a copy

Abstract

Virus influenza A atau dikenal dengan avian influenza virus (AIV) merupakan virus yang digolongkan dalam famili orthomyxoviridae. Tingginya tingkat mutasi dan besarnya kemampuan segmen gen virus influenza A untuk melakukan penyusunan ulang menghasilkan berbagai macam varian virus influenza A dengan karakterisitik, transmisi dan potensi ancaman terhadap manusia yang berbeda-beda. Karakter dan cara transmisi yang dihasilkan oleh varian virus influenza A dapat dilihat dari struktur proteinnya yaitu dengan cara menganalisis struktur sekunder protein virus influenza A. Sedemikian hingga, dengan analisis lebih lanjut maka obat antivirus maupun tes diagnosa dapat dirancang untuk menghambat penyebaran virus influenza A. Pentingnya struktur sekunder protein sebagai objek dalam analisis, maka penelitian ini melakukan prediksi struktur sekunder protein virus influenza A berbasis model machine learning menggunakan ekstraksi fitur vectors of hydrophobic, steric, and electronic properties (VHSE) dan parameter konformasi sebagai pengkode asam amino serta sliding window untuk pembentukan pola. Dari peneltian ini didapatkan bahwa hasil prediksi sturktur sekunder protein virus influenza A dengan model random forest menggunakan VHSE dan parameter konformasi sebagai pengkode asam amino memiliki kemiripan (similarity) sebesar 93,87% terhadap struktur sekunder protein virus influenza A yang sebenarnya.
=================================================================================================================================
Influenza A virus or known as avian influenza virus (AIV) is a virus that classified in the orthomyxoviridae family. The high rate of mutation and the ability of influenza A virus gene segments to rearrange produce various influenza A virus variants with different characteristics, transmission and potential threats to humans. The character and mode of transmission produced by the influenza A virus variant can be seen from its protein structure, by analyzing the secondary structure of the influenza A virus protein. Thus, with further analysis, antiviral drugs and diagnostic tests can be designed to inhibit the spread of the influenza A virus. The importance of protein secondary structure as an object in analysis, so this research predicts the secondary structure of influenza A virus proteins based on machine learning models using feature extraction vectors of hydrophobic, steric, and electronic properties (VHSE) and conformation parameters as an amino acid encoder with sliding window for pattern formation. From this research, it was found that the results of predicting the secondary structure of the influenza A virus protein byrandom forest model using VHSE and the conformation parameters as encoding for amino acids had a similarity of 93,87% to the real secondary structure of the influenza virus protein.

Item Type: Thesis (Masters)
Uncontrolled Keywords: model machine learning, parameter konformasi, sliding window, struktur sekunder protein, VHSE, virus influenza A, conformation parameter, influenza A virus, machine learning model, protein secondary structure
Subjects: Q Science > Q Science (General) > Q325.5 Machine learning. Support vector machines.
Q Science > QR Microbiology > QR355 Virology
Divisions: Faculty of Science and Data Analytics (SCIENTICS) > Mathematics > 44101-(S2) Master Thesis
Depositing User: Sanada Aulia Fanani
Date Deposited: 08 Aug 2024 06:06
Last Modified: 11 Sep 2024 02:50
URI: http://repository.its.ac.id/id/eprint/113619

Actions (login required)

View Item View Item