During early stages of drug R&D, it is important to select targets genes with high disease relevance and generate hypotheses to support the relationship. Many pharmaceutical companies, drug discovery ventures, and members of academia spend a great deal of time, effort and expense to identify for target genes and generate hypotheses of potential targets, but identifying highly novel target genes that can be first in class continues to be a difficult task. Discovery and hypothesis generation becomes even more difficult, for highly novel targets where the relationship between the disease and the target gene has not been explicitly described in literature.

In the white paper, FRONTEO uses KIBIT, an AI engine that discovers unreported relationships from published literature, to analyze textual data published in over 600 leading journals held by Springer Nature. KIBIT provides specific examples to explain the process of finding unreported target genes with high disease relevance from gene networks and generates hypotheses regarding the unreported relationships.