Skip to content
Back to outputs

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

Research output: Contribution to journal › Article › peer-review

Standard

A consensus neural network-based technique for discriminating soluble and poorly soluble compounds. / Manallack, D.; Tehan, B.; Gancia, E.; Hudson, Brian; Ford, M.; Livingstone, D.; Whitley, David; Pitt, W.

In: Journal of Chemical Information and Computer Sciences, Vol. 43, No. 2, 2003, p. 674-679.

Research output: Contribution to journal › Article › peer-review

Harvard

Manallack, D, Tehan, B, Gancia, E, Hudson, B, Ford, M, Livingstone, D, Whitley, D & Pitt, W 2003, 'A consensus neural network-based technique for discriminating soluble and poorly soluble compounds', Journal of Chemical Information and Computer Sciences, vol. 43, no. 2, pp. 674-679. https://doi.org/10.1021/ci0202741

APA

Manallack, D., Tehan, B., Gancia, E., Hudson, B., Ford, M., Livingstone, D., Whitley, D., & Pitt, W. (2003). A consensus neural network-based technique for discriminating soluble and poorly soluble compounds. Journal of Chemical Information and Computer Sciences, 43(2), 674-679. https://doi.org/10.1021/ci0202741

Vancouver

Manallack D, Tehan B, Gancia E, Hudson B, Ford M, Livingstone D et al. A consensus neural network-based technique for discriminating soluble and poorly soluble compounds. Journal of Chemical Information and Computer Sciences. 2003;43(2):674-679. https://doi.org/10.1021/ci0202741

Author

Manallack, D. ; Tehan, B. ; Gancia, E. ; Hudson, Brian ; Ford, M. ; Livingstone, D. ; Whitley, David ; Pitt, W. / A consensus neural network-based technique for discriminating soluble and poorly soluble compounds. In: Journal of Chemical Information and Computer Sciences. 2003 ; Vol. 43, No. 2. pp. 674-679.

Bibtex

@article{f1284c4be42c4acab876726a94de332a,
title = "A consensus neural network-based technique for discriminating soluble and poorly soluble compounds",
abstract = "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.",
author = "D. Manallack and B. Tehan and E. Gancia and Brian Hudson and M. Ford and D. Livingstone and David Whitley and W. Pitt",
year = "2003",
doi = "10.1021/ci0202741",
language = "English",
volume = "43",
pages = "674--679",
journal = "Journal of Chemical Information and Computer Sciences",
issn = "0095-2338",
publisher = "American Chemical Society",
number = "2",

}

RIS

TY - JOUR

T1 - A consensus neural network-based technique for discriminating soluble and poorly soluble compounds

AU - Manallack, D.

AU - Tehan, B.

AU - Gancia, E.

AU - Hudson, Brian

AU - Ford, M.

AU - Livingstone, D.

AU - Whitley, David

AU - Pitt, W.

PY - 2003

Y1 - 2003

N2 - 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.

AB - 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.

U2 - 10.1021/ci0202741

DO - 10.1021/ci0202741

M3 - Article

VL - 43

SP - 674

EP - 679

JO - Journal of Chemical Information and Computer Sciences

JF - Journal of Chemical Information and Computer Sciences

SN - 0095-2338

IS - 2

ER -

ID: 228030