Please login first
Blood-Brain Barrier Passage Prediction Using Decision Tree
* 1, 2 , 1, 2 , 3 , 4 , 5
1  Unidad de Toxicología Experimental, Universidad de Ciencias Médicas de Villa Clara, Santa Clara, Villa Clara, Cuba. CP: 50200, Cuba
2  Bioinformatic Research in Systems & Computer Engineering, Carleton University, Ottawa, Canada
3  School of Medicine and Pharmacy Vietnam National University, Hanoi Y1 Building, 144 Xuan Thuy, Cau Giay, Hanoi
4  Hanoi University of Pharmacy, 13-15 Le Thanh Tong, Hoan Kiem, Hanoi, Viet-nam. dSchool of Medicine and Pharmacy, Vietnam National University, Hanoi (VNU) 144 Xuan Thuy, Cau Giay, Hanoi, Viet-nam
5  Departamento de Química, Universidade Federal de Lavras, CP 3037, 37200-000, Lavras, MG, Brazil


In this report, the blood brain barrier (BBB) permeability prediction is carried out using a decision tree. A recently published data set of 497 compounds is selected to develop the tree model. The developed model shows an accuracy of 87.66% for training set; 86.09% in the 10-fold cross-validation procedure and 87.93% for the test set. Some structural explanation of how our model describe the passage of molecules through the BBB is given. Moreover, a comparison with other approaches is carried out showing good behaviour of our method. Finally, we can say that, the present results could represent a useful tools available and reproducible by all scientific community in the early stages of neuropharmaceutical drug discovery/development projects.

Keywords: Blood-brain barrier; Classification tree; Molecular descriptor; Quantitative Structure-Activity Relationship; WEKA