There are major differences between the structures of human dihydrofolate reductase (hDHFR) and Mycobacterium tuberculosis dihydrofolate reductase (mtDHFR). These differences may allow the design of more selective mtDHFR inhibitors. In this paper, we have used docking approaches to study the binding orientations and predict binding affinities of 2,4-diamino-5-methyl-5-deazapteridines derivatives in both hDHFR and mtDHFR. Our results of molecular docking combined with experimental data for inhibition of the human and mycobacterial dihydrofolate reductases suggest the presence of empty spaces around the 2,4-diaminodeazapteridine and N10-phenyl rings in the mtDHFR active site that are not found in the hDHFR-bound structures. Preparation of new analogs with substituents attached to C7 of the pteridine nucleus and positions 3 and 4 of the N10-phenyl group should increase the affinity and selectivity for mtDHFR.