Abstract
Cancer comprises a set of more than 200 diseases resulting from the uncontrolled growth of cells that invade tissues and organs, which can spread to other regions of the body. The types of cancer found in humans are also described in animal models, a fact that has raised the interest of the scientific community in comparative oncology studies. In this study, bioinformatics tools were used to implement a computational model that uses text mining and natural language processing to construct a reference database that relates human and canine genes potentially associated with cancer, defining genetic pathways and information about cancer and cancer therapies. The CANCROX reference database was constructed by processing the scientific literature and lists more than 1300 drugs and therapies used to treat cancer, in addition to over 10 000 combinations of these drugs, including 40 types of cancer. A user-friendly interface was developed that enables researchers to search for different types of information about therapies, drug combinations, genes and types of cancer. In addition, data visualization tools allow to explore and relate different drugs and therapies for the treatment of cancer, providing information for groups studying animal models, in this case the dog, as well as groups studying cancer in humans.