This study presents a computational method to predict new molecular targets for known drugs, using chemical similarity between drugs and ligand sets. The approach, called the Similarity Ensemble Approach (SEA), compares drug-target interactions based on ligand similarity rather than target sequence or structure. The method was applied to 3,665 FDA-approved and investigational drugs and 65,241 ligands from the MDL Drug Data Report (MDDR) database, resulting in 901,590 drug-target comparisons. The SEA approach identified 6,928 drug-ligand pairs with significant similarity, with expectation values (E-values) better than 1 × 10⁻¹⁰. These predictions were validated against known associations and unreported polypharmacology. For example, the drug DMT was found to bind multiple serotonin receptors, including 5-HT₁A, 5-HT₁B, 5-HT₁D, 5-HT₂A, 5-HT₂B, 5-HT₂C, 5-HT₅A, and 5-HT₆, with affinities ranging from 39 nM to 2.1 μM. This supports the role of 5-HT₂A as the primary target for DMT's hallucinogenic effects. The study also identified several new drug-target associations, including the binding of the drug Prozac to the β₁ adrenergic receptor and the drug Vadilex to the 5-HT transporter. These findings highlight the importance of off-target effects in drug action, which can contribute to therapeutic efficacy or adverse effects. For instance, the drug Motilium was found to bind to α₁ adrenergic receptors, which may contribute to its cardiovascular side effects. Similarly, the drug Prozac was found to bind to the β₁ adrenergic receptor, which may explain its cardiovascular effects. The study also identified drugs that bind to targets unrelated by sequence or structure to their canonical targets, such as the reverse transcriptase inhibitor Rescriptor binding to the histamine H₄ receptor. These findings demonstrate the potential of the SEA approach to identify new drug-target interactions and to understand the physiological relevance of off-target effects. The study concludes that the SEA approach is a systematic and comprehensive method for predicting drug-target interactions, which can help in the discovery of new drug indications and the understanding of drug side effects.This study presents a computational method to predict new molecular targets for known drugs, using chemical similarity between drugs and ligand sets. The approach, called the Similarity Ensemble Approach (SEA), compares drug-target interactions based on ligand similarity rather than target sequence or structure. The method was applied to 3,665 FDA-approved and investigational drugs and 65,241 ligands from the MDL Drug Data Report (MDDR) database, resulting in 901,590 drug-target comparisons. The SEA approach identified 6,928 drug-ligand pairs with significant similarity, with expectation values (E-values) better than 1 × 10⁻¹⁰. These predictions were validated against known associations and unreported polypharmacology. For example, the drug DMT was found to bind multiple serotonin receptors, including 5-HT₁A, 5-HT₁B, 5-HT₁D, 5-HT₂A, 5-HT₂B, 5-HT₂C, 5-HT₅A, and 5-HT₆, with affinities ranging from 39 nM to 2.1 μM. This supports the role of 5-HT₂A as the primary target for DMT's hallucinogenic effects. The study also identified several new drug-target associations, including the binding of the drug Prozac to the β₁ adrenergic receptor and the drug Vadilex to the 5-HT transporter. These findings highlight the importance of off-target effects in drug action, which can contribute to therapeutic efficacy or adverse effects. For instance, the drug Motilium was found to bind to α₁ adrenergic receptors, which may contribute to its cardiovascular side effects. Similarly, the drug Prozac was found to bind to the β₁ adrenergic receptor, which may explain its cardiovascular effects. The study also identified drugs that bind to targets unrelated by sequence or structure to their canonical targets, such as the reverse transcriptase inhibitor Rescriptor binding to the histamine H₄ receptor. These findings demonstrate the potential of the SEA approach to identify new drug-target interactions and to understand the physiological relevance of off-target effects. The study concludes that the SEA approach is a systematic and comprehensive method for predicting drug-target interactions, which can help in the discovery of new drug indications and the understanding of drug side effects.