4D-QSAR analysis and pharmacophore modeling: Electron conformational-genetic algorithm approach for penicillins

Yanmaz E., SARIPINAR E., Sahin K., Gecen N., Copur F.

BIOORGANIC & MEDICINAL CHEMISTRY, cilt.19, sa.7, ss.2199-2210, 2011 (SCI-Expanded) identifier identifier identifier

  • Yayın Türü: Makale / Tam Makale
  • Cilt numarası: 19 Sayı: 7
  • Basım Tarihi: 2011
  • Doi Numarası: 10.1016/j.bmc.2011.02.035
  • Dergi Adı: BIOORGANIC & MEDICINAL CHEMISTRY
  • Derginin Tarandığı İndeksler: Science Citation Index Expanded (SCI-EXPANDED), Scopus
  • Sayfa Sayıları: ss.2199-2210
  • Anahtar Kelimeler: Penicillins, 4D-QSAR, Drug design, Pharmacophore, Electron conformational, Genetic algorithm, MOLECULAR-ORBITAL THEORY, BIOACTIVITY PREDICTION, VARIABLE SELECTION, BETA-LACTAMS, QSAR, IDENTIFICATION, BINDING, THIOSEMICARBAZONE, ANTIBIOTICS, ORIENTATION
  • Erciyes Üniversitesi Adresli: Evet

Özet

4D-QSAR studies were performed on a series of 87 penicillin analogues using the electron conformational-genetic algorithm (EC-GA) method. In this EC-based method, each conformation of the molecular system is described by a matrix (ECMC) with both electron structural parameters and interatomic distances as matrix elements. Multiple comparisons of these matrices within given tolerances for high active and low active penicillin compounds allow one to separate a smaller number of matrix elements (ECSA) which represent the pharmacophore groups. The effect of conformations was investigated building model 1 and 2 based on ensemble of conformers and single conformer, respectively. GA was used to select the most important descriptors and to predict the theoretical activity of the training (74 compounds) and test (13 compounds, commercial penicillins) sets. The model 1 for training and test sets obtained by optimum 12 parameters gave more satisfactory results (R-training(2) = 0.861, SEtraining = 0.044, R-test(2) = 0.892, SEtest = 0.099, q(2) = 0.702, q(ext1)(2) = 0.777 and q(ext2)(2) = 0.733) than model 2 (R-training(2) = 0.774, SEtraining = 0.056, R-test(2) = 0.840, SEtest = 0.121, q(2) = 0.514, q(ext1)(2) = 0.641 and q(ext2)(2) = 0.570). To estimate the individual influence of each of the molecular descriptors on biological activity, the E statistics technique was applied to the derived EC-GA model. (C) 2011 Elsevier Ltd. All rights reserved.

4D-QSAR studies were performed on a series of 87 penicillin analogues using the electron conformational-genetic algorithm (EC-GA) method. In this EC-based method, each conformation of the molecular system is described by a matrix (ECMC) with both electron structural parameters and interatomic distances as matrix elements. Multiple comparisons of these matrices within given tolerances for high active and low active penicillin compounds allow one to separate a smaller number of matrix elements (ECSA) which represent the pharmacophore groups. The effect of conformations was investigated building model 1 and 2 based on ensemble of conformers and single conformer, respectively. GA was used to select the most important descriptors and to predict the theoretical activity of the training (74 compounds) and test (13 compounds, commercial penicillins) sets. The model 1 for training and test sets obtained by optimum 12 parameters gave more satisfactory results (R(training)(2) = 0.861, SE(training) = 0.044, R(test)(2) = 0.892, SE(test) = 0.099, q(2) = 0.702, q(ext1)(2) = 0.777 and q(ext2)(2) = 0.733) than model 2 (R(training)(2) = 0.774, SE(training) = 0.056, R(test)(2) = 0.840, SE(test) = 0.121, q(2) = 0.514, q(ext1)(2) = 0.641 and q(ext2)(2) = 0.570). To estimate the individual influence of each of the molecular descriptors on biological activity, the E statistics technique was applied to the derived EC-GA model. (C) 2011 Elsevier Ltd. All rights reserved.