Utilization of Graph Theory Modelling in Searching Potential Indonesian Medicine Plants as Antihypertension

Abstract

Hypertension can be defined as a disease characterized by a systolic blood pressure value of at least 140 mmHg and or a diastolic blood pressure value of at least 90 mmHg. Broadly speaking, hypertension is divided into 2 types. Tissue Pharmacology, based on the concept of multidisciplinary integrity, is a powerful tool for analyzing multi-level tissue from molecular-target-pathways-of-disease through interactions between traditional medicine and disease from a holistic perspective The materials used in this study include: search for proteins related to hypertension at Uniprot, download results of interaction designs of compounds and proteins on STRING, cluster results from the Cytoscape application, download results of potential substance data on KNApSAcK, results of data downloads information on substances in PubChem, results of RStudio processing, results of data searches, extraction and labeling from Jupyter Lab, results of downloading of interaction designs of compounds and proteins on STITCH, which will later be analyzed and combined using the Orange application. The search process for significant proteins related to hypertension is carried out in the Uniprot database on the initial web display. There is a display of protein search results related to hypertension select on the left that reads "Homo sapiens" because the search for hypertension protein is only limited to proteins found in humans. The calculation of overall centrality is done by entering a script/command in the kernel in the Python program. The results obtained are that the AGTR1_HUMAN protein has an overall centrality value of (2.76967082423431), the GNAS_HUMAN protein has an overall centrality value of ( -1,38194807757488), ACE2_HUMAN protein has an overall centrality value of (-1,72856432286345), TAC3_HUMAN protein has an overall centrality value (-1.01621749330442), EDN1_HUMAN protein has an overall centrality value of (-0.392822941980431). These values indicate how significant a protein is in a cluster. The results of the search for the best 1 plant found 5 plants that can bind 50% of the total protein including: Euphorbia hirta which has 3 compounds with 4 target proteins and has a percentage of the target protein is 80%, Mangifera indica which has 2 compounds with a target protein of 4 proteins and has a percentage of the target protein is 80%, Capsicum frutescens which has 3 compounds with a target protein of 3 proteins and has a percentage of the target protein is 60%, Carica papaya which has 3 compounds with 3 target proteins and has a percentage of the target protein is 60%, and Trigonella foenum graecum which has 1 compound with 3 target proteins and has a percentage of the target protein is 60%. In the combination of 2 plants, 14 formulas were obtained where all of these formulas targeted all the protein interactions obtained, so that the percentage value for all proteins was 100%.