Penerapan Graph Clustering Pada Jaringan Homogen Kesamaan Senyawa Obat Untuk Identifikasi Kandidat Drug Repurposing Penyakit Alzheimer

Nafisah, Aghnina Azka Nafisah (2026) Penerapan Graph Clustering Pada Jaringan Homogen Kesamaan Senyawa Obat Untuk Identifikasi Kandidat Drug Repurposing Penyakit Alzheimer. Other thesis, Institut Teknologi Sepuluh Nopember.

Text 5002221116-Undergraduate_Thesis.pdf - Accepted Version Restricted to Repository staff only Download (23MB) | Request a copy

Abstract

Penyakit Alzheimer merupakan penyakit neurodegeneratif progresif yang menjadi penyebab utama demensia di dunia. Seiring dengan sifat penyakit yang terus berkembang dan menyebabkan penurunan fungsi kognitif secara bertahap, diperlukan upaya untuk menangani penyakit ini sejak dini guna memperlambat progresivitas dan meningkatkan kualitas hidup pasien. Oleh karena itu, penemuan obat baru yang efektif untuk terapi Alzheimer masih menjadi kebutuhan penting dalam bidang kesehatan. Salah satu pendekatan yang dapat digunakan untuk mempercepat proses tersebut adalah drug repurposing. Penelitian ini bertujuan mengidentifikasi kandidat drug repurposing Alzheimer menggunakan pendekatan graph clustering berbasis data interaksi obat–target protein dari DrugBank. Data obat berstatus approved ditransformasikan menjadi drug similarity network menggunakan metrik Jaccard similarity dengan threshold 0,3. Jaringan yang terbentuk terdiri atas 2.232 simpul dan 18.091 sisi. Struktur komunitas dalam jaringan kemudian diidentifikasi menggunakan algoritma Louvain yang menghasilkan 189 komunitas dengan nilai modularitas 0,9143. Klaster yang mengandung obat Alzheimer resmi, yaitu Donepezil, Galantamine, Rivastigmine, dan Memantine, digunakan sebagai klaster acuan untuk proses identifikasi kandidat. Seleksi bertahap berdasarkan keanggotaan klaster, hubungan langsung dengan obat referensi, dan nilai Jaccard similarity menghasilkan 7 kandidat drug repurposing dari 174 kandidat awal. Validasi komputasional menggunakan molecular docking dengan AutoDock Vina menunjukkan bahwa Benzgalantamine memiliki afinitas pengikatan terbaik terhadap protein asetilkolinesterase (1EVE) dengan nilai binding affinity sebesar -12,056 kcal/mol, diikuti oleh Demecarium sebesar -8,644 kcal/mol. Hasil ini menunjukkan bahwa pendekatan graph clustering berbasis jaringan kesamaan senyawa obat efektif dalam mengidentifikasi kandidat drug repurposing yang berpotensi untuk terapi penyakit Alzheimer.
==================================================================================================================================
Alzheimer's disease is a progressive neurodegenerative disorder and the leading cause of dementia worldwide. Due to its progressive nature and gradual decline in cognitive function, efforts are needed to address the disease at an early stage in order to slow its progression and improve patients' quality of life. Therefore, the discovery of new and effective therapeutic agents for Alzheimer's disease remains an important challenge in the healthcare field. One approach that can accelerate this process is drug repurposing. This study aims to identify potential drug repurposing candidates for Alzheimer's disease using a graph clustering approach based on drug–target protein interaction data obtained from DrugBank. Approved drugs were transformed into a drug similarity network using the Jaccard similarity metric with a threshold of 0.3. The resulting network consisted of 2,232 nodes and 18,091 edges. Community structures within the network were then identified using the Louvain algorithm, resulting in 189 communities with a modularity value of 0.9143. Clusters containing approved Alzheimer's drugs, namely Donepezil, Galantamine, Rivastigmine, and Memantine, were used as reference clusters for candidate identification. A stepwise selection process based on cluster membership, direct connections to reference drugs, and Jaccard similarity values yielded 7 drug repurposing candidates from an initial set of 174 candidates. Computational validation using molecular docking with AutoDock Vina showed that Benzgalantamine exhibited the strongest binding affinity toward acetylcholinesterase protein (PDB ID: 1EVE) with a binding affinity value of -12.056 kcal/mol, followed by Demecarium with a binding affinity value of -8.644 kcal/mol. These findings indicate that a graph clustering approach based on a drug similarity network is effective for identifying potential drug repurposing candidates for Alzheimer's disease therapy.

Item Type:	Thesis (Other)
Uncontrolled Keywords:	Alzheimer, Interaksi Obat-Target Protein, Drug Repurposing, Graph Clustering, Algoritma Louvain Alzheimer’s disease, Drug–Target Protein Interaction, Drug Repurposing, Graph Clustering, Louvain Algorithm
Subjects:	Q Science > Q Science (General) > Q325.5 Machine learning. Support vector machines. Q Science > QA Mathematics > QA278.55 Cluster analysis
Divisions:	Faculty of Science and Data Analytics (SCIENTICS) > Mathematics > 44201-(S1) Undergraduate Thesis
Depositing User:	Aghnina Azka Nafisah
Date Deposited:	29 Jul 2026 04:16
Last Modified:	29 Jul 2026 04:17
URI:	http://repository.its.ac.id/id/eprint/139453

Actions (login required)

View Item