N-Myristoyltransferase (NMT) is an attractive target for the development of novel antimalarial drugs. In this paper, we report a hologram quantitative structure-activity relationship (HQSAR), a comparative molecular field analysis (CoMFA), and a comparative molecular similarity indices analysis (CoMSIA) studies on a series of benzo[b]thiophene-containing selective inhibitors of plasmodial NMT. Two QSAR models were constructed, one for Plasmodium falciparum NMT and another for Homo sapiens NMT, for each QSAR method. The predictive ability of all QSAR models was significant (r2 > 0.90 and q2 > 0.70). Comparing results of different enzymes, we achieved great understanding of structure-selectivity relationship. Finally, these QSAR models …show more content…
∑▒〖(Y_predicted- Y_observed)〗^2 is the predictive residual sum of squares (PRESS) and the lowest PRESS value is used to derive the final PLS model.
RESULTS AND DISCUSSION
HQSAR Analysis
HQSAR was performed on 39 benzo[b]thiophene derivatives using three distinct parameters (fragment size, hologram length, and fragment type). Initially, 10 HQSAR models were generated using the different fragment distinction with the fragment size 5–8. The combination of atom, connection, donor and acceptor gave the best model for both enzymes (q2 = 0.783, r2 = 0.971, SEE = 0.521 for PfNMT; q2 = 0.770, r2 = 0.964, SEE = 0.533 for HsNMT) based on the hologram length of 293 and 5 components.
HQSAR contribution map analysis
The HQSAR results gave direct evidence about the individual atomic contributions to the biological activity through the use of different color codes. The contribution of different fragments for the activity of molecules 7, 23, and 36 are displayed in Fig. 1A–C, respectively. The colors at the red end of the spectrum indicates the poor contributions (red and orange), while colors at the green end reflect favorable contributions (yellow and