Target Baru Pengobatan Meningitis berdasarkan Centrality Measure jaringan protein dan Self Oganizing Map.

Meningitis is a serious threat to health with potentially fatal consequences. Understanding protein interactions related to chronic conditions is crucial for the development of effective treatments. In silico analysis is considered to have greater effectiveness because it simulates through computation and tries various possibilities at a lower cost. This study aims to analyze protein-protein interactions related to Meningitis with cluster analysis techniques on undirected graphs. The proposed method is the Self Organizing Map (SOM) algorithm as a cluster. This algorithm can cluster undirected graph-based protein interaction data. Protein data involved in Meningitis disease comes from OMIM. From this data, proteins belonging to the gene locus are explored for their interactions, resulting in interaction data in the form of an undirected graph. The combination of centrality measure is used for feature engineering on undirected graph data. The main protein candidates are potentially located in the Cluster 1 model with the largest silhouette score (0.359) and Davies-Bouldin Index (1.667). The cluster has 18 proteins with the highest significance to Meningitis. From the overall centrality ranking results, the three highest significance proteins are CISH (3.921222), TNFSF10 (3.403541), and ICAM3 (2.623702) which have the potential to become Meningitis target proteins. CISH protein has the highest overall centrality score value compared to the others, so CISH protein may be a new alternative in the treatment of Meningitis.