The Janus-associated kinase 2 (JAK2) V617F mutation is believed to play a critical role in the pathogenesis of polycythemia vera, essential thrombocythemia, and idiopathic myelofibrosis. The discovery of activating mutations associated with inhibition of cascade of events mediated by JAK2 target became an attractive approach for the treatment of myeloproliferative disorder. In this study, we performed a ligand-based pharmacophore modeling to explore the important chemical features of JAK2 inhibitors. The top ten hypotheses were generated based on 47 known inhibitors of JAK2 using PHASE module of Schrodinger software. The best pharmacophore hypothesis was found to be AADDR.212 which consists of two acceptors, two donors and one ring aromatic group. The selected model was validated by survival score, selectivity, and GH score. Two types of validation studies were done which includes potency validation by virtual screening against set of decoys, and selectivity validation by screening against set of inhibitors of JAK1, JAK2, JAK3, and Tyk2 (all tyrosine kinase family proteins). The selected model was utilized as a 3D query to screen against ZINC natural and chemical database, and subsequently the screened compounds were filtered by applying the Lipinski’s rule of five, ADME properties and molecular docking. Finally, fifteen compounds were obtained as novel virtual hits to inhibit the JAK2 enzyme.