Combining Unsupervised and Supervised Artificial Neural Networks to Predict Aquatic Toxicity. Gini, G., Craciun, M., Konig, C., & Benfenati, E. J.~Chem.~Inf.~Model., 44(6):1897--1902, 2004. ![Combining Unsupervised and Supervised Artificial Neural Networks to Predict Aquatic Toxicity [link]](https://bibbase.org/img/filetypes/link.svg "link")
Paper abstract bibtex Abstract: Most quantitative structure-activity relationship (QSAR) models are linear relationships and significant for only a limited domain of compounds. Here we propose a data-driven approach with a flexible combination of unsupervised and supervised neural networks able to predict the toxicity of a large set of different chemicals while still respecting the QSAR postulates. Since QSAR is applicable only to similar compounds, which have similar biological and physicochemical properties, large numbers of compounds are clustered before building local models, and local models are ensembled to obtain the final result. The approach has been used to develop models to predict the fish toxicity of Pimephales promelas and Tetrahymena pyriformis, a protozoan.
@article{Gini:2004aa,
Abstract = {Abstract: Most quantitative structure-activity relationship (QSAR) models are linear relationships and significant for only a limited domain of compounds. Here we propose a data-driven approach with a flexible combination of unsupervised and supervised neural networks able to predict the toxicity of a large set of different chemicals while still respecting the QSAR postulates. Since QSAR is applicable only to similar compounds, which have similar biological and physicochemical properties, large numbers of compounds are clustered before building local models, and local models are ensembled to obtain the final result. The approach has been used to develop models to predict the fish toxicity of Pimephales promelas and Tetrahymena pyriformis, a protozoan.},
Affiliation = {Istituto di Ricerche Farmacologiche "Mario Negri", Via Eritrea 62, 20157, Milano, Italy},
Author = {Gini, G. and Craciun, M.V. and Konig, C. and Benfenati, E.},
Date-Added = {2007-12-11 17:01:03 -0500},
Date-Modified = {2008-04-28 14:13:42 -0400},
Issn = {1549--9596},
Journal = {J.~Chem.~Inf.~Model.},
Number = {6},
Pages = {1897--1902},
Title = {Combining Unsupervised and Supervised Artificial Neural Networks to Predict Aquatic Toxicity},
Url = {http://pubs3.acs.org/acs/journals/doilookup?in_doi=10.1021/ci0401219},
Volume = {44},
Year = {2004},
Bdsk-Url-1 = {http://pubs3.acs.org/acs/journals/doilookup?in_doi=10.1021/ci0401219}}
Downloads: 0
{"_id":"gnLhFp7jWyRxoGn9Y","bibbaseid":"gini-craciun-konig-benfenati-combiningunsupervisedandsupervisedartificialneuralnetworkstopredictaquatictoxicity-2004","downloads":0,"creationDate":"2016-02-18T13:03:34.752Z","title":"Combining Unsupervised and Supervised Artificial Neural Networks to Predict Aquatic Toxicity","author_short":["Gini, G.","Craciun, M.","Konig, C.","Benfenati, E."],"year":2004,"bibtype":"article","biburl":"https://dl.dropboxusercontent.com/u/26998770/main.bib","bibdata":{"bibtype":"article","type":"article","abstract":"Abstract: Most quantitative structure-activity relationship (QSAR) models are linear relationships and significant for only a limited domain of compounds. Here we propose a data-driven approach with a flexible combination of unsupervised and supervised neural networks able to predict the toxicity of a large set of different chemicals while still respecting the QSAR postulates. Since QSAR is applicable only to similar compounds, which have similar biological and physicochemical properties, large numbers of compounds are clustered before building local models, and local models are ensembled to obtain the final result. The approach has been used to develop models to predict the fish toxicity of Pimephales promelas and Tetrahymena pyriformis, a protozoan.","affiliation":"Istituto di Ricerche Farmacologiche \"Mario Negri\", Via Eritrea 62, 20157, Milano, Italy","author":[{"propositions":[],"lastnames":["Gini"],"firstnames":["G."],"suffixes":[]},{"propositions":[],"lastnames":["Craciun"],"firstnames":["M.V."],"suffixes":[]},{"propositions":[],"lastnames":["Konig"],"firstnames":["C."],"suffixes":[]},{"propositions":[],"lastnames":["Benfenati"],"firstnames":["E."],"suffixes":[]}],"date-added":"2007-12-11 17:01:03 -0500","date-modified":"2008-04-28 14:13:42 -0400","issn":"1549--9596","journal":"J.~Chem.~Inf.~Model.","number":"6","pages":"1897--1902","title":"Combining Unsupervised and Supervised Artificial Neural Networks to Predict Aquatic Toxicity","url":"http://pubs3.acs.org/acs/journals/doilookup?in_doi=10.1021/ci0401219","volume":"44","year":"2004","bdsk-url-1":"http://pubs3.acs.org/acs/journals/doilookup?in_doi=10.1021/ci0401219","bibtex":"@article{Gini:2004aa,\n\tAbstract = {Abstract: Most quantitative structure-activity relationship (QSAR) models are linear relationships and significant for only a limited domain of compounds. Here we propose a data-driven approach with a flexible combination of unsupervised and supervised neural networks able to predict the toxicity of a large set of different chemicals while still respecting the QSAR postulates. Since QSAR is applicable only to similar compounds, which have similar biological and physicochemical properties, large numbers of compounds are clustered before building local models, and local models are ensembled to obtain the final result. The approach has been used to develop models to predict the fish toxicity of Pimephales promelas and Tetrahymena pyriformis, a protozoan.},\n\tAffiliation = {Istituto di Ricerche Farmacologiche \"Mario Negri\", Via Eritrea 62, 20157, Milano, Italy},\n\tAuthor = {Gini, G. and Craciun, M.V. and Konig, C. and Benfenati, E.},\n\tDate-Added = {2007-12-11 17:01:03 -0500},\n\tDate-Modified = {2008-04-28 14:13:42 -0400},\n\tIssn = {1549--9596},\n\tJournal = {J.~Chem.~Inf.~Model.},\n\tNumber = {6},\n\tPages = {1897--1902},\n\tTitle = {Combining Unsupervised and Supervised Artificial Neural Networks to Predict Aquatic Toxicity},\n\tUrl = {http://pubs3.acs.org/acs/journals/doilookup?in_doi=10.1021/ci0401219},\n\tVolume = {44},\n\tYear = {2004},\n\tBdsk-Url-1 = {http://pubs3.acs.org/acs/journals/doilookup?in_doi=10.1021/ci0401219}}\n\n","author_short":["Gini, G.","Craciun, M.","Konig, C.","Benfenati, E."],"key":"Gini:2004aa","id":"Gini:2004aa","bibbaseid":"gini-craciun-konig-benfenati-combiningunsupervisedandsupervisedartificialneuralnetworkstopredictaquatictoxicity-2004","role":"author","urls":{"Paper":"http://pubs3.acs.org/acs/journals/doilookup?in_doi=10.1021/ci0401219"},"downloads":0},"search_terms":["combining","unsupervised","supervised","artificial","neural","networks","predict","aquatic","toxicity","gini","craciun","konig","benfenati"],"keywords":[],"authorIDs":[],"dataSources":["c5japf9eAQRaeMS4h"]}