A consensus neural network-based technique for discriminating soluble and poorly soluble compounds

被引:56
作者
Manallack, DT
Tehan, BG
Gancia, E
Hudson, BD
Ford, MG
Livingstone, DJ
Whitley, DC
Pitt, WR
机构
[1] Celltech R&D Ltd, Cambridge CB1 6GS, England
[2] Monash Univ, Victorian Coll Pharm, Dept Med Chem, Parkville, Vic, Australia
[3] Univ Portsmouth, Ctr Mol Design, Portsmouth PO1 2DY, Hants, England
[4] ChemQuest, Sandown PO36 8LZ, Wight, England
来源
JOURNAL OF CHEMICAL INFORMATION AND COMPUTER SCIENCES | 2003年 / 43卷 / 02期
关键词
D O I
10.1021/ci0202741
中图分类号
O6 [化学];
学科分类号
0703 ;
摘要
BCUT [Burden, CAS, and University of Texas] descriptors, defined as eigenvalues of modified connectivity matrices, have traditionally been applied to drug design tasks such as defining receptor relevant subspaces to assist in compound selections. In this paper we present studies of consensus neural networks trained on BCUTs to discriminate compounds with poor aqueous solubility from those with reasonable solubility. This level was set at 0.1 mg/mL on advice from drug formulation and drug discovery scientists. By applying strict criteria to the insolubility predictions, approximately 95% of compounds are classified correctly. For compounds whose predictions have a lower level of confidence, further parameters are examined in order to flag those considered to possess unsuitable biopharmaceutical and physicochemical properties. This approach is not designed to be applied in isolation but is intended to be used as a filter in the selection of screening candidates, compound purchases, and the application of synthetic priorities to combinatorial libraries.
引用
收藏
页码:674 / 679
页数:6
相关论文
共 41 条
[1]   The correlation and prediction of the solubility of compounds in water using an amended solvation energy relationship [J].
Abraham, MH ;
Le, J .
JOURNAL OF PHARMACEUTICAL SCIENCES, 1999, 88 (09) :868-880
[2]  
[Anonymous], NETLAB ALGORITHMS PA
[3]  
Bishop C. M., 1995, NEURAL NETWORKS PATT
[4]   Search for predictive generic model of aqueous solubility using Bayesian neural nets [J].
Bruneau, P .
JOURNAL OF CHEMICAL INFORMATION AND COMPUTER SCIENCES, 2001, 41 (06) :1605-1616
[5]   MOLECULAR-IDENTIFICATION NUMBER FOR SUBSTRUCTURE SEARCHES [J].
BURDEN, FR .
JOURNAL OF CHEMICAL INFORMATION AND COMPUTER SCIENCES, 1989, 29 (03) :225-227
[6]  
Clark DE, 2001, COMB CHEM HIGH T SCR, V4, P477
[7]  
Furka A, 2002, DRUG DISCOV TODAY, V7, P1
[8]   Application of BCUT metrics and genetic algorithm in binary QSAR analysis [J].
Gao, H .
JOURNAL OF CHEMICAL INFORMATION AND COMPUTER SCIENCES, 2001, 41 (02) :402-407
[9]   A knowledge-based approach in designing combinatorial or medicinal chemistry libraries for drug discovery. 1. A qualitative and quantitative characterization of known drug databases [J].
Ghose, AK ;
Viswanadhan, VN ;
Wendoloski, JJ .
JOURNAL OF COMBINATORIAL CHEMISTRY, 1999, 1 (01) :55-68
[10]   Strategic pooling of compounds for high-throughput screening [J].
Hann, M ;
Hudson, B ;
Lewell, X ;
Lifely, R ;
Miller, L ;
Ramsden, N .
JOURNAL OF CHEMICAL INFORMATION AND COMPUTER SCIENCES, 1999, 39 (05) :897-902