Non-linear Prediction of Quantitative Structure-Activity Relationships

Peter Tino; I Nabney; BS Williams; J Losel; Y Sun

doi:10.1021/ci034255i

Non-linear Prediction of Quantitative Structure-Activity Relationships

Peter Tino, I Nabney, BS Williams, J Losel, Y Sun

Computer Science

Research output: Contribution to journal › Article

19 Citations (Scopus)

Abstract

Predicting the log of the partition coefficient P is a long-standing benchmark problem in Quantitative Structure-Activity Relationships (QSAR). In this paper we show that a relatively simple molecular representation (using 14 variables) can be combined with leading edge machine learning algorithms to predict logP on new compounds more accurately than existing benchmark algorithms which use complex molecular representations.

Original language	English
Pages (from-to)	1647-1653
Number of pages	7
Journal	Journal of Chemical Information and Computer Sciences
Volume	44
Issue number	5
DOIs	https://doi.org/10.1021/ci034255i
Publication status	Published - 1 Jan 2004

Access to Document

10.1021/ci034255i

Cite this

@article{b37be6244c444aeebd497a1d7058d43f,

title = "Non-linear Prediction of Quantitative Structure-Activity Relationships",

abstract = "Predicting the log of the partition coefficient P is a long-standing benchmark problem in Quantitative Structure-Activity Relationships (QSAR). In this paper we show that a relatively simple molecular representation (using 14 variables) can be combined with leading edge machine learning algorithms to predict logP on new compounds more accurately than existing benchmark algorithms which use complex molecular representations.",

author = "Peter Tino and I Nabney and BS Williams and J Losel and Y Sun",

year = "2004",

month = jan,

day = "1",

doi = "10.1021/ci034255i",

language = "English",

volume = "44",

pages = "1647--1653",

journal = "Journal of Chemical Information and Computer Sciences",

publisher = "American Chemical Society",

number = "5",

}

TY - JOUR

T1 - Non-linear Prediction of Quantitative Structure-Activity Relationships

AU - Tino, Peter

AU - Nabney, I

AU - Williams, BS

AU - Losel, J

AU - Sun, Y

PY - 2004/1/1

Y1 - 2004/1/1

N2 - Predicting the log of the partition coefficient P is a long-standing benchmark problem in Quantitative Structure-Activity Relationships (QSAR). In this paper we show that a relatively simple molecular representation (using 14 variables) can be combined with leading edge machine learning algorithms to predict logP on new compounds more accurately than existing benchmark algorithms which use complex molecular representations.

AB - Predicting the log of the partition coefficient P is a long-standing benchmark problem in Quantitative Structure-Activity Relationships (QSAR). In this paper we show that a relatively simple molecular representation (using 14 variables) can be combined with leading edge machine learning algorithms to predict logP on new compounds more accurately than existing benchmark algorithms which use complex molecular representations.

UR - http://www.scopus.com/inward/record.url?scp=5444224314&partnerID=8YFLogxK

U2 - 10.1021/ci034255i

DO - 10.1021/ci034255i

M3 - Article

C2 - 15446822

VL - 44

SP - 1647

EP - 1653

JO - Journal of Chemical Information and Computer Sciences

JF - Journal of Chemical Information and Computer Sciences

IS - 5

ER -

Non-linear Prediction of Quantitative Structure-Activity Relationships

Abstract

Access to Document

Fingerprint

Cite this