Support vector machines for drug discovery

Heikamp, Kathrin; Bajorath, Jürgen

doi:10.1517/17460441.2014.866943

Cited by 148 publications

(104 citation statements)

References 74 publications

Supporting

Mentioning

104

Contrasting

Order By: Relevance

“…This method can be used to solve both classification and regression problems. We used the SVM embedded in “e1071” package from R, invoked through R statistics module in Pipeline Pilot 8.535. According to reported literatures, SVM are among the best-performing approaches for chemical and biological property prediction and the computational identification of active compounds35.…”

Section: Methodsmentioning

confidence: 99%

“…We used the SVM embedded in “e1071” package from R, invoked through R statistics module in Pipeline Pilot 8.535. According to reported literatures, SVM are among the best-performing approaches for chemical and biological property prediction and the computational identification of active compounds35. SVM projects the data into a higher dimensional feature space where linear separation is frequently possible, facilitating object classification, ranking and regression-based property value prediction.…”

Section: Methodsmentioning

confidence: 99%

See 1 more Smart Citation

Predicting Subtype Selectivity for Adenosine Receptor Ligands with Three-Dimensional Biologically Relevant Spectrum (BRS-3D)

Kuang

et al. 2016

Sci Rep

View full text Add to dashboard Cite

Adenosine receptors (ARs) are potential therapeutic targets for Parkinson’s disease, diabetes, pain, stroke and cancers. Prediction of subtype selectivity is therefore important from both therapeutic and mechanistic perspectives. In this paper, we introduced a shape similarity profile as molecular descriptor, namely three-dimensional biologically relevant spectrum (BRS-3D), for AR selectivity prediction. Pairwise regression and discrimination models were built with the support vector machine methods. The average determination coefficient (r2) of the regression models was 0.664 (for test sets). The 2B-3 (A2B vs A3) model performed best with q2 = 0.769 for training sets (10-fold cross-validation), and r2 = 0.766, RMSE = 0.828 for test sets. The models’ robustness and stability were validated with 100 times resampling and 500 times Y-randomization. We compared the performance of BRS-3D with 3D descriptors calculated by MOE. BRS-3D performed as good as, or better than, MOE 3D descriptors. The performances of the discrimination models were also encouraging, with average accuracy (ACC) 0.912 and MCC 0.792 (test set). The 2A-3 (A2A vs A3) selectivity discrimination model (ACC = 0.882 and MCC = 0.715 for test set) outperformed an earlier reported one (ACC = 0.784). These results demonstrated that, through multiple conformation encoding, BRS-3D can be used as an effective molecular descriptor for AR subtype selectivity prediction.

show abstract

Section: Methodsmentioning

confidence: 99%

Section: Methodsmentioning

confidence: 99%

Predicting Subtype Selectivity for Adenosine Receptor Ligands with Three-Dimensional Biologically Relevant Spectrum (BRS-3D)

Kuang

et al. 2016

Sci Rep

View full text Add to dashboard Cite

show abstract

“…However, the other three contributions are derived as weighted sums from the support vectors, in which not only λ (i) but also the different denominators contribute to the weighting. In the example shown, λ (3) has consistently higher weights than λ (1) or λ (2) .…”

Section: ■ Introductionmentioning

confidence: 85%

“…9,25,26 SVMs utilizing kernels usually have much higher prediction capacity than linear models. 1 However, the use of kernel functions comes at the price of black box character and lack of model interpretability. If the explicit mapping ϕ(x) is not available, it is impossible to determine contributions of the features to the classification.…”

Section: ■ Introductionmentioning

confidence: 99%

Visualization and Interpretation of Support Vector Machine Activity Predictions

Balfer

Bajorath

2015

J. Chem. Inf. Model.

Self Cite

View full text Add to dashboard Cite

Support vector machines (SVMs) are among the preferred machine learning algorithms for virtual compound screening and activity prediction because of their frequently observed high performance levels. However, a well-known conundrum of SVMs (and other supervised learning methods) is the black box character of their predictions, which makes it difficult to understand why models succeed or fail. Herein we introduce an approach to rationalize the performance of SVM models based upon the Tanimoto kernel compared with the linear kernel. Model comparison and interpretation are facilitated by a visualization technique, making it possible to identify descriptor features that determine compound activity predictions. An implementation of the methodology has been made freely available.

show abstract

“…Relatively recent publications have provided some in-depth discussions regarding small sample estimation [4][5][6][7][8], where fuzzy clustering and support vector machine (SVM) have received special attentions [9][10][11]. Fuzzy clustering and SVM have been applied to address various problems through progression as the methodologies themselves advance, such as classification, regression, image classification, human activity, geo-marketing analysis, and drug discovery [12][13][14][15][16][17][18][19]. More importantly, recent studies have provided additional insights regarding SVM [20][21][22].…”

Section: Introductionmentioning

confidence: 99%

Methodologies for assessing costs of rail transit systems based on small sample data

Wang

Zhang

Chen

et al. 2015

International Journal of Rail Transportation

View full text Add to dashboard Cite

China has developed plans to build 87 mass transit rail lines, totalling 2500 km, in 25 cities from 2009 to 2015. The life-cycle costs of the urban rail transit systems have become the focus of both the government and the private sector involved in these large-scale investments. However, the availability of quality data has posed a major challenge to such life-cycle cost analyses; in other words, for any methodology to be effective, it must have the capability of working with very limited amount of available data, or small sample data. In this article, two cost assessment methodologies, fuzzy cluster and support vector machine, are proposed to analyse the life-cycle cost of urban rail transit systems based on small sample data. A case study featuring Line 1 of the Shijiazhuang urban rail transit system was employed to demonstrate the validity of the proposed methodologies. The analysis results indicate that the two assessment methodologies are valid for the life-cycle cost assessment of urban rail transit systems when only small sample data are available.

show abstract

Support vector machines for drug discovery

Cited by 148 publications

References 74 publications

Predicting Subtype Selectivity for Adenosine Receptor Ligands with Three-Dimensional Biologically Relevant Spectrum (BRS-3D)

Predicting Subtype Selectivity for Adenosine Receptor Ligands with Three-Dimensional Biologically Relevant Spectrum (BRS-3D)

Visualization and Interpretation of Support Vector Machine Activity Predictions

Methodologies for assessing costs of rail transit systems based on small sample data

Contact Info

Product

Resources

About