23Drug repurposing, identifying novel indications for drugs, bypasses common drug 24 development pitfalls to ultimately deliver therapies to patients faster. However, 25 most repurposing discoveries have been led by anecdotal observations (e.g. 26Viagra) or experimental-based repurposing screens, which are costly, time-27 consuming, and imprecise. Recently, more systematic computational 28 approaches have been proposed, however these rely on utilizing the information 29 from the diseases a drug is already approved to treat. This inherently limits the 30 algorithms, making them unusable for investigational molecules. Here, we 31 present a computational approach to drug repurposing, CATNIP, that requires 32 only biological and chemical information of a molecule. CATNIP is trained with 33 2,576 diverse small molecules and uses 16 different drug similarity features, 34 such as structural, target, or pathway based similarity. This model obtains 35 significant predictive power (AUC = 0.841). Using our model, we created a 36repurposing network to identify broad scale repurposing opportunities between 37 drug types. By exploiting this network, we identified literature-supported 38repurposing candidates, such as the use of systemic hormonal preparations for 39