Fig. 1

Computational repositioning identifies a polo-like kinase 1/2 inhibitor as a candidate drug for pulmonary fibrosis. (A) Schematic of the in-silico approach used to identify candidate drugs that attenuate pulmonary fibrosis. Microarray data of lung tissue of patients with idiopathic pulmonary fibrosis (IPF) were extracted from the Gene Expression Omnibus database. By comparing them with microarray data from healthy donors, the differentially expressed genes (DEGs) were established. Therapeutic candidate drugs that antagonize the DEG signature were identified using drug signatures obtained from the Library of integrated Network-based Cellular Signatures. (B) Drug-disease score represents the agreement between the input genetic variation data and the candidate drug’s genetic variation data. The top 50 candidate drugs with the highest drug-disease scores when the fold-change (FC) of DEGs was |log₂FC| > 0.585 (> 1.5 and < 0.66) are shown. (C) The top 50 candidate drugs with the highest drug-disease scores when FC of DEGs is |log₂FC| > 1 (> 2.0 and < 0.5) are shown. The blue triangle points toward nintedanib, which was approved for the treatment of IPF. The red triangle points toward BI2536, a polo-like kinase 1/2 inhibitor that was selected for a therapeutic candidate of IPF