Comparison of Support Vector Machine and Artificial Neural Network Systems for Drug/Nondrug Classification

By Evgeny Byvatov, Uli Fechner, J. Sadowski, Gisbert Schneider. Journal of chemical information and computer sciences, 2003

567 citations recorded by the source corpus.

Subfields: Computational Theory and Mathematics, Molecular Biology, Spectroscopy

View Paper: https://www.semanticscholar.org/paper/e874fdcccdf49a47d8436819b1b15b17506f7bf5

How researchers see this paper

The researchers whose trusted networks rank it highest, strongest advocate first. Every researcher gets a different order.

In Computational Theory and Mathematics

1. Sepp Hochreiter
2. Thomas Unterthiner
3. George Karypis
4. Günter Klambauer
5. Nikil Wale
6. Michihiro Kuramochi
7. Andreas Zell
8. Andreas Mayr

In Molecular Biology

1. Sepp Hochreiter
2. Thomas Unterthiner
3. Günter Klambauer
4. Jean-Philippe Vert
5. Laurent Jacob
6. Mehmet Gönen
7. Andreas Mayr
8. Michihiro Kuramochi

In Spectroscopy

1. George Karypis
2. Nikil Wale
3. Michihiro Kuramochi
4. Andreas Zell
5. Andreas Mayr
6. Mehmet Gönen
7. Holger Fröhlich
8. Günter Klambauer

Citations in the corpus

• Protein-ligand interaction prediction: an improved chemogenomics approach (2008, background): “We test this strategy on three important classes of drug targets, namely enzymes, G-protein-coupled receptors (GPCR) and ion channels, and report dramatic improvements in prediction accuracy over classical ligandbased virtual screening, in particular for targets with few or no known ligands.”
• Toxicity Prediction using Deep Learning (2015, methodology): “…scoring approaches like the Naive Bayes statistics (Xia et al., 2004; Nigsch et al., 2008; Mussa et al., 2013), density estimation (R. et al., 2012; Harper et al., 2001), nearest neighbor, support vector machines, and shallow feed forward neural networks (Byvatov et al., 2003; Lowe et al., 2011).”
• Towards Optimal Descriptor Subset Selection with Support Vector Machines in Classification and Regression (2004, methodology): “The usefulness of SVMs in drug design has for instance recently been shown by Byvatov et al. [4] [5].”
• Frequent substructure-based approaches for classifying chemical compounds (2003, methodology): “This includes techniques based on principle component regression and partial least squares [81], neural networks [5], [27], [57], [85], recursive partitioning [4], [18], [72], phylogenetic-like trees [64], [78], binary QSAR [30], [51], linear discriminant analysis [67], and support vector machines [”
• A Systematic Prediction of Multiple Drug-Target Interactions from Chemical, Genomic, and Pharmacological Data (2012, background): “Theoretically, LBVS compares candidate ligands with the known drugs of a target protein to find new compounds using statistical tools [8,9].”
• Deep Learning in Medical Imaging (2019, methodology): “13 Logistic regression parameterizes the logistic model to predict binary classification.”
• A Structure-Based Drug Discovery Paradigm (2019, methodology): “SVMs with supervised ML algorithms use binary activity prediction to distinguish between a drug and nondrug [98,99] or between specific and nonspecific compounds [100,101].”
• Frequent sub-structure-based approaches for classifying chemical compounds (2005, methodology): “This includes techniques based on principle component regression and partial least squares [81], neural networks [5], [27], [57], [85], recursive partitioning [4], [18], [72], phylogenetic-like trees [64], [78], binary QSAR [30], [51], linear discriminant analysis [67], and support vector machines [”
• Towards Multimodal Sarcasm Detection (An _Obviously_ Perfect Paper) (2019, background): “SVMs are strong predictors for small-sized datasets and at times outperform neural counterparts (Byvatov et al., 2003).”