CARLSBAD is a cheminformatics platform that provides researchers with the ability to generate novel hypotheses pertaining to the relationships between biological targets, chemical compounds and their common chemical patterns (CCPs). This capability has implications for lead compound prediction, drug repurposing, target identification, off-target prediction and more. Below are three examples to help illustrate the power of using CARLSBADs cheminformatics algorithms and complex network theory to relate CCPs to confederated bioactivity data in order to generate potential lead compounds for a disease-related target, Caspase 8, discover the potential target of a widely-used drug for which the mechanism of action remains elusive, Praziquantel, and to predict possible off-targets for a drug, Rosiglitazone.

Example 1: Lead Compound Generation

In order to develop new therapeutics for cancers and neurodegenerative diseases, CARLSBAD was used to discover potential lead compounds for Caspase 8. Caspase 8 is a cysteine-aspartic acid protease that plays a central role in Apoptosis. In addition, the cleaved form plays a role in the formation of the autophagosome, an autophagic vacuole. Thus, Caspase 8 is a promising target for modulation of both cell death and organelle recycling. Using CARLSBAD, we were able to quickly identify more than 200 lead compounds that potentially will modulate Caspase 8. Experimental studies are currently underway to examine the activity between Caspase 8 and these compounds.

Example 2: Target Identification

Schistosomiasis is a parasitic disease infecting more than 200 million people in over 70 countries. In addition, another 800 million people live in areas where they are at risk of infection. Schistosomiasis is implicated in more than 200,000 deaths annually and is responsible for more than 50 million disability-adjusted life years (DALYs). Praziquantel (PZQ) has been the only widely used drug to treat schistosomiasis for more than 30 years. Despite its wide spread usage, PZQs mechanism of action (MOA) remains unknown. With nearly 1/6th of the global population infected or living in regions where they are at risk of infection, the consequences of drug resistant strains of schistosomes emerging would be dire. Hence, the need to identify PZQs MOA is crucial in order to rationally develop alternative treatments to PZQ. With CARLSBAD, we were able to easily identify more than 50 compounds structurally similar to PZQ that all targeted the same protein. This protein exists in schistosomes and there is reason to believe, based on what is known about schistosomes treated with PZQ, that it is an integral part of PZQs MOA.

Example 3: Off-Target Predication

Undesired side-effects due to off-target protein interactions keep many therapeutics from being commercially developed. In addition, off-target interactions can become known after a drug has been commercialized and this can lead to personal harm and costly litigation. Thus there is a need to better predict off-target interactions. With CARLSBAD we examined off-target interactions for Rosiglitazone. Rosiglitazone is an antidiabetic thiazolidinedione. It's mechanism works by binding PPAR receptors in fat cells and making them more responsive to insulin. Since being approved by the FDA, concerns over reports of the drug's association with increased risk of heart attacks has led to tens of thousands of lawsuits. Using CARLSBAD, we identified 24 previously unidentified off-targets of Rosiglitazone, 10 of which have a direct role in cardiac function. Thus, the ability to better predict off-target interactions might lead to a greater number of safe therapies being commercially developed.