Workshop 3 :

HyperChem Software from Biological Active Molecules and SAR to QSAR Modeling.

Dr. Rafik Menacer (CRSP)

Dr. Saad BOUCHEKIOUA (CRSP)

Abstract:

The computationally predicted active molecules need to be tested experimentally. Therefore, according to the classical methods, the effect of different parameters on the activity in question requires carrying experiments. Hence a huge number of experiments could be conducted in addition to an excessive number of trials especially when several factors are considered. For this purpose, researchers described a new method to optimize the number of experiments which allows obtaining as much information as possible on the activity/parameters dependency. This technique is noted as the experimental design methodology.

Outline

I. HyperChem software features and calculations: Dr. Saad BOUCHEKIOUA (CRSP)

a) Drawing and editing basicsb) Molecular Mechanics Calculationsc) Quantum Mechanics Calculationsd) Chirality Modelinge) Conformational searchf) Molecules geometrical measurementsg) Molecular structure properties


II. QSAR methodology and the related statistics: Dr. Rafik MENACER (CRSP)

a) Virtual screening of active molecules series

b) SAR patternsc) Molecular descriptors generation and selectiond) Molecular optimization
  1. Semiempirical Methods
  2. Quantum based molecular descriptors

e) QSAR Modeling
  1. Data statistical treatment
  2. Multi Linear Modeling MLR
  3. Polynomial Modeling
  4. QSAR model validation and statistical analysis