ReDEx is a comprehensive tool featuring multiple analysis tabs that allow users to explore individual drugs, diseases, or specific drug–disease pairs. It supports the analysis of single or multiple SMILES strings, identifying the most similar drugs using the Tanimoto Coefficient (TC). ReDEx also facilitates drug repurposing by integrating disease-gene networks with drug similarity metrics. Additionally, it enables users to investigate shared targets between drugs and diseases and perform enrichment analysis to uncover deeper biological insights. - SMILES Similarity Check - Drug Repurpose - Intersection SMILES Similarity Check: You can choose to input SMILES either manually or by uploading an Excel file. Provide one or multiple SMILES to retrieve, for each, the most similar drugs based on the Tanimoto coefficient. The resulting table will include the following columns for each input SMILE: TC: Tanimoto coefficient; DrugName_Link: Name of the similar drug with its associated link; DrugID: ID of the similar drug; Associated_Disease_Link: If the similar drug has any association with a disease along with its associated link; Drug_DiseaseIDs: The disease ID (if exist) that the similar drug had any relationship with; Drug_Phase: The phase (if exist) of the similar drug and its respective disease; Drug_Smile: Smile of the similar drug; Input_Smile: The smile from the input data. Drug Repurpose: The dataset includes 3,275 diseases and in the parenthesis is the number of approved or investigational drugs for the respective disease. Once the user chooses one or multiple diseases, the tool runs network and structure based algorithms to find scores, combine them and calculate a final score. The higher the Final Score of a drug the most candidate it is to be repurposed for the disease. Intersection Analysis: For the exploration of common target genes, the user can choose either loading files or data using our built-in datasets. If the user wishes to load files then is mandatory for both files, to have a column named 'EntrezID' which has the entrezIDs and a column named 'Symbol' which has the Gene names. Once the data are loaded and run, a table is being produced showing the common genes between the drug and the disease modules loaded. In this table, the column Disease_Drug indicates the disease(s) and drug(s) in which each common gene is found. Right below, a Venn Diagram is also produced showing the intersection of the number of genes and it's significance. The p-value is calculated using the hypergeometric distribution test. If multiple diseases and/or multiple drugs are selected from our built-in dataset, a heatmap will be generated for each group: one for diseases and/or one for drugs, depending on which group includes more than one item. Each heatmap displays the Jaccard similarity among the selected items. Drug-Disease Network: The tool calculates several network algorithms that are also used in Cytohubba in Cytoscape with one extra called TrustRank. First a table is produced showing what genes were not found in the Gene-Gene Interaction dataset. Then a second table is produced showing the interactions between the common genes found. This table can also be saved and used in Cytoscape from the user. Right below this table, the Disease-Drug Network is shown using the common genes. The bigger the dot the more connections there are. Next, there is a table with all the scores calculated in their normalized values. This is done because the CRank aggregated score is computed based on the normalized values of each algorithm. If the user wishes to see only one, there is the option with the dropdown menu above the plot, showing each algorithm separately, with it's initial scores. Simultanousley a relevant network plot is shown based on the algorithm and the top N genes to show in the network. It is worth noting here that the genes shown are the ones that have connections. Those that don't are simply discarded to reduce noice. So for example if common genes are 100 and you see here less, is because some don't have any connections at all with the common genes. Enrichment Analysis: Each time the analysis runs, users can choose from different categories: ALL – All categories together: BP, MF, CC. Good choice to start for a broad analysis. BP – Biological Process MF – Molecular Function CC – Cellular Component KEGG – Kyoto Encyclopedia of Genes and Genomes Additionally, various p-adjustment methods are available, defining the p-adjusted value. The first table highlights overrepresented pathways using the common genes. Below, another table shows gene abundance—how often a gene appears in significant pathways. Users can visualize this with a bar plot, selecting specific genes (default: first 5). A dot plot follows, displaying the top 15 significant pathways (y-axis) and their GeneRatio (x-axis). Dot color represents the p-adjusted value, while size indicates the number of genes from the comparison. Next, a cnetplot illustrates the top 3 most statistically significant pathways and their gene symbols. Lastly, an emapplot shows the top 15 pathways and their relationships based on shared genes, with dot color indicating the p-adjusted value and size representing gene count.
