Molecular Similarity Searching Using Atom Environments, Information-Based Feature Selection, and a Naïve Bayesian Classifier

By Andreas Bender, H. Mussa, Robert C. Glen, S. Reiling. 2004

241 citations recorded by the source corpus.

Subfields: Computational Theory and Mathematics, Molecular Biology, Spectroscopy

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. Antony J. Williams
2. Igor V. Tetko
3. Stefano Forli
4. Sepp Hochreiter
5. Thomas Unterthiner
6. Günter Klambauer
7. David S. Goodsell
8. Igor I. Baskin

In Molecular Biology

1. Sepp Hochreiter
2. Thomas Unterthiner
3. Stefano Forli
4. Günter Klambauer
5. David S. Goodsell
6. Antony J. Williams
7. Arthur J. Olson
8. Igor V. Tetko

In Spectroscopy

1. Igor V. Tetko
2. Antony J. Williams
3. Noel M. O'Boyle
4. Geoffrey R. Hutchison
5. Chris Morley
6. Andreas Zell
7. Andreas Mayr
8. Andreas Bender

Citations in the corpus

• Interpreting linear support vector machine models with heat map molecule coloring (2011, methodology): “Common molecular encodings that generate interpretable features are radial atom environments [3], depth first search paths [15], or extended connectivity fingerprints (ECFP) [16].”
• Open Babel: An open chemical toolbox (2011, methodology): “…a molecule; the fingerprint format [13] and fastsearch format [14] are used for similarity and substructure searching (see below); the MolPrint2D and Multilevel Neighborhoods of Atoms formats calculate circular fingerprints defined by Bender et al. [15,16] and Filimonov et al. [17,18] respectively.”
• DeepTox: Toxicity Prediction using Deep Learning (2016, methodology): “Since the arithmetic mean is not a reasonable choice for combining the predictions of different models , DeepTox uses a probabilistic approach with similar assumptions as naive Bayes ( see Supplementary Section 3 ) to fully exploit the probabilistic predictions in our ensembles .”
• Virtual Screening with AutoDock: Theory and Practice (2010, methodology): “We were encouraged that this MolPrint2D cutoff (>10) allowed us to also retain compounds that were only vaguely similar to the known inhibitors.”
• Molecular fingerprint similarity search in virtual screening. (2015, background): “◦ Molprint2D [22,23]: Molprint2D encodes the atom environments of each atom of the molecular connectivity table, which are represented by strings of varying size.”
• Online chemical modeling environment (OCHEM): web platform for data storage, model development and publishing of chemical information (2011, methodology): “1 descriptors [46], MERA and MerSy descriptors [47–50], MolPrint 2D descriptors [51], ShapeSignatures [52] and logP and aqueous solubility calculated with ALOGPS program [24].”