شماره ركورد كنفرانس :
3976
عنوان مقاله :
QSAR study for inhibitory activities of Phosphodiesterase
پديدآورندگان :
Baher Elham Golestan University , Khosravi Zahra z.khosravi972@gmail.com Golestan University , Gharaghani Sajjad Depatrment of Bioinformatics Biophisics, Tehran University, Tehran, Iran
كليدواژه :
Quantitative structure–activity relationship (QSAR) , Inhibitory activity , Phosphodiesterase, IC50
عنوان كنفرانس :
ششمين سمينار ملي دوسالانه كمومتريكس ايران
چكيده فارسي :
A quantitative structure-activity relationship (QSAR) study has been done on the
inhibitory activity of some phosphodiesterase. QSAR models were based on multiple
linear regression (MLR), genetic algorithm (GA) and successive projection algorithm
(SPA). First, stepwise MLR was used as a linear descriptor selection procedure and then
SPA and GA was employed as a non-linear descriptor selection method. Finally,
selected descriptors were employed for constructing model by supported vector
regression (SVR) and MLR techniques. Comparison of the statistical results after cross
validation was indicated the priority of nonlinear methods to MLR models.