شماره ركورد كنفرانس :
3976
عنوان مقاله :
Quantitative structure activity relationship study of quinazoline derivatives as tyrosine kinase (EGFR) inhibitors using scaled conjugate gradient artificial neural network
پديدآورندگان :
Beglari M. Biglari55@yahoo.com Shahrood University of Technology , Goudarzi N. Shahrood University of Technology , Shahsavani D. Shahrood University of Technology , Arab Chamjangali M. Shahrood University of Technology , Dosti R. Shahrood University of Technology
كليدواژه :
Quantitative structure activity relationship (QSAR) , Quinazoline derivatives , Epidermal growth factor receptor (EGFR) , Docking , Artificial neural network (ANN)
عنوان كنفرانس :
ششمين سمينار ملي دوسالانه كمومتريكس ايران
چكيده فارسي :
Tyrosine kinases are important mediators of signal transduction process, leading to
programmed cell death. Epidermal growth factor receptor (EGFR) which plays a vital
role as a regulator of cell growth is one of the intensely studied tyrosine kinase targets
of inhibitors. EGFR is overexpressed in many human cancers including non-small cell
lung cancer, bladder cancer, and breast cancer. Since the hyper-activation of EGFR has
been associated with these diseases, inhibitor of EGFR has potential therapeutic value
and it has been extensively studied in the pharmaceutical industry [1, 2].
A nonlinear quantitative structure activity relationship study was presented for modeling
and predicting epidermal growth factor receptor (EGFR) inhibitor data. Scaled
conjugate gradient artificial neural network (SCG-ANN) was used to link molecular
structures and inhibitory data. A data set consisting of 43 derivatives of analogues of
quinazoline was used in this study [3]. Among a large number of calculated molecular
descriptors by Dragon software, only six significant molecular descriptors were
obtained by stepwise regression. The selected descriptors were combined with two (E2
and logKi) as newly docking derived descriptors and then they were used as inputs for
neural network. The neural network architecture and its parameters were optimized. The
prediction ability of the model was evaluated using the test set. The mean square errors
and mean absolute errors for the test set data were 0.6934 and 0.6591, respectively. The
results obtained showed the excellent prediction ability and stability of the proposed
model in the prediction of inhibitory activity data of the corresponding analogues.