Quantitative structure–activity relationship study using genetic algorithm–enhanced replacement method combined with molecular docking studies of isatin derivatives as inhibitors of human transglutaminase 2

Inhibition of the enzyme human transglutaminase 2 (TG2) is important for the treatment of various diseases. Isatin derivatives, including 3‐acylidene‐2‐oxoindoles, inhibit of human TG2. In this work, quantitative structure–activity relationship (QSAR) and molecular docking studies were performed on the inhibitory activity of isatin derivatives. First, genetic algorithm (GA) was used for selecting the sum of descriptors. Then a small number of descriptors were selected using the enhanced replacement method (ERM) and the stepwise (SW) method. The descriptors nCl, HE, R4m+, and n = CHR were selected using GA‐ERM and the descriptors R6m, R5e +, n = CHR, and HE were produced using the GA‐SW method. Two linear and nonlinear methods, namely multiple linear regression (MLR) and support vector regression (SVR), were used for QSAR modeling. GA‐ERM outperformed in combination with the SVR method, which indicated more power at the prediction of inhibitory activity. Also, molecular docking studies were performed for understanding the mechanism of interaction of isatin derivatives with TG2 and investigating the effect of some descriptors such as the hydration energy (HE) on the inhibitory activity of the studied compounds. Docking results revealed that the formation of hydrogen bonding and hydrophobic interactions play important roles in the connection between the protein TG2 (PDB: 3S3P) and isatin derivatives.