Graph Theory Enabled Drug Repurposing
One challenge with human designed pharmaceutical drugs for specific illnesses is the risk of unwanted drug interactions and side effects. Media for popular versions these drugs often include dense warnings of these effects and dangers, but less is heard of the other aspect of unintended drug functions: additional curative properties. The specific active mechanisms in these engineered products vary so slightly that their precise interactions with most scenarios cannot be known from laboratory testing and theory.
Using the known bio-molecular interactions of similar drugs and graph theory, Ruggero Gramatica et. al explore the additional beneficial interactions in their paper published on plosone.org, “Graph Theory Enables Drug Repurposing – How a Mathematical Model Can Drive the Discovery of Hidden Mechanisms of Action.” By scanning other scientific papers with a computer algorithm (this action of finding linked papers to specific topics uses graph theory in itself) and combining the relevant information of drug interactions into a large scale graph, they can formulate meaningful suggestions into hidden mechanisms of pharmaceutical drugs and treatments. Using this graph and statistical methods the researchers succeeded in finding these secondary effects, and successfully treated rare illnesses, such as the granulomatous disease Sarcoidosis.
Through this network of drug functions, if expanded, doctors could find alternative drugs for a patient that cannot take the usual drug, or, as discussed in the paper, to find a combination of drugs to attack a rare disease or illness with secondary drug effects. This graph system connects to the fundamentals of a large scale graph as discussed in lecture with complex components and bridges. The authors then used path algorithms to rank the connections with tie strength and path length assisting in ranking of effectiveness. With a magnitude scale on the bonds, they could categorize the strength of bonds between drugs and their effects. Until humans can design drugs without side effects and more powerful, specific actions, wide scope analysis of the side effects will minimize risks when consuming pharmaceuticals. Over time, ideally, the graph of connections will contain more nodes with fewer connecting edges as scientists design more niche drugs.
http://www.plosone.org/article/info%3Adoi%2F10.1371%2Fjournal.pone.0084912