Comparison of topological descriptors for similarity-based virtual screening using multiple bioactive reference structures

Hert, Jérôme; Willett, Peter; Wilton, David J.; Acklin, Pierre; Azzaoui, Kamal; Jacoby, Edgar; Schuffenhauer, Ansgar

doi:10.1039/b409865j

Cited by 245 publications

(308 citation statements)

References 31 publications

(71 reference statements)

Supporting

Mentioning

294

Contrasting

Order By: Relevance

“…19 CATS descriptors were calculated using an in-house Pipeline Pilot protocol; the histogram was not normalized by the number of heavy atoms as reported in the implementation description; 16 instead, the occurrence of each pharmacophore pair was binned into an 8-bit string. 19 The FEPOPS descriptors were generated using a Novartis in-house program. The Tanimoto coefficient was used for all similarity calculations.…”

Section: Similarity Measuresmentioning

confidence: 99%

Quantifying the Relationships among Drug Classes

Hert

Keiser

Irwin

et al. 2008

J. Chem. Inf. Model.

151

165

View full text Add to dashboard Cite

The similarity of drug targets is typically measured using sequence or structural information. Here, we consider chemo-centric approaches that measure target similarity on the basis of their ligands, asking how chemoinformatics similarities differ from those derived bioinformatically, how stable the ligand networks are to changes in chemoinformatics metrics, and which network is the most reliable for prediction of pharmacology. We calculated the similarities between hundreds of drug targets and their ligands and mapped the relationship between them in a formal network. Bioinformatics networks were based on the BLAST similarity between sequences, while chemoinformatics networks were based on the ligand-set similarities calculated with either the Similarity Ensemble Approach (SEA) or a method derived from Bayesian statistics. By multiple criteria, bioinformatics and chemoinformatics networks differed substantially, and only occasionally did a high sequence similarity correspond to a high ligand-set similarity. In contrast, the chemoinformatics networks were stable to the method used to calculate the ligand-set similarities and to the chemical representation of the ligands. Also, the chemoinformatics networks were more natural and more organized, by network theory, than their bioinformatics counterparts: ligand-based networks were found to be small-world and broad-scale.

show abstract

Section: Similarity Measuresmentioning

confidence: 99%

Quantifying the Relationships among Drug Classes

Hert

Keiser

Irwin

et al. 2008

J. Chem. Inf. Model.

151

165

View full text Add to dashboard Cite

show abstract

“…In this respect, we note that the hologram representation involves a superimposed coding procedure that results in each element of the fingerprint encoding information about multiple linear substructures, and each linear substructure contributing to the occurrences in multiple fingerprint elements. This many-to-many mapping is very different from the one-toone mapping represented by the Sunset fingerprints and the near one-to-one mapping represented by the ECFC_4 fingerprints (where the hashing to 1024 elements introduces a very limited degree of fragment overlap [10]). Tables 4 and 5.…”

Section: Full Resultsmentioning

confidence: 91%

“…The Pipeline Pilot ECFC_4 fingerprints encode circular substructures describing a central atom and all atoms within a two-bond radius of it. These substructures are processed using a hashing scheme based on the Morgan algorithm for graph canonicalisation [46]; in our experiments, the resulting integer codes were again hashed to give a fixed-length fingerprint containing 1024 elements, a procedure that we have found to be highly effective in similarity-based virtual screening experiments [10,47]. The two types of descriptor hence differ in both the types of topological substructure that are encoded (linear or circular, for holograms and ECFC_4, respectively) and in the number of fingerprint-elements associated with each fragment (three or one, for holograms and ECFC_4, respectively); however both types of substructure are highly specific in nature and the resulting fingerprints hence provide a very precise description of molecular topology.…”

Section: Structural Representationsmentioning

confidence: 99%

“…A 2D fingerprint is a vector that encodes the presence or absence of topological substructures (typically atom-, bond-or ring-centred fragments) in a molecule [7,8], and many different types of fingerprint have been described in the literature [9,10]. A fingerprint is clearly an extremely simple type of structural representation, but still contains sufficient information to enable effective similarity-based virtual screening to be carried out (see, e.g., [11][12][13][14][15][16][17]).…”

Section: Introductionmentioning

confidence: 99%

See 1 more Smart Citation

Analysis and use of fragment-occurrence data in similarity-based virtual screening

Arif

Holliday

Willett

2009

J Comput Aided Mol Des

Self Cite

View full text Add to dashboard Cite

Current systems for similarity-based virtual screening use similarity measures in which all the fragments in a fingerprint contribute equally to the calculation of structural similarity. This paper discusses the weighting of fragments on the basis of their frequencies of occurrence in molecules. Extensive experiments with sets of active molecules from the MDL Drug Data Report and the World of Molecular Bioactivity databases, using fingerprints encoding Tripos holograms, Pipeline Pilot ECFC_4 circular substructures and Sunset Molecular keys, demonstrate clearly that frequency-based screening is generally more effective than conventional, unweighted screening. The results suggest that standardising the raw occurrence frequencies by taking the square root of the frequencies will maximise the effectiveness of virtual screening. An upper-bound analysis shows the complex interactions that can take place between representations, weighing schemes and similarity coefficients when similarity measures are computed, and provides a rationalisation of the relative performance of the various weighting schemes.

show abstract

“…9,10 It is evident, that the RTR fulfills both goals of VS validation as stated above: comparability of methods and estimation of the expected hit number. Often so-called "enrichment factors" (EF) 11 are calculated from the RTR, that are meant to normalize to the null hypothesis of uniformly distributed active molecules in the final ranking list.…”

Section: Introductionmentioning

confidence: 87%

Impact of Benchmark Data Set Topology on the Validation of Virtual Screening Methods: Exploration and Quantification by Spatial Statistics

Rohrer

Baumann

2008

J. Chem. Inf. Model.

View full text Add to dashboard Cite

A common finding of many reports evaluating ligand-based virtual screening methods is that validation results vary considerably with changing benchmark data sets. It is widely assumed that these data set specific effects are caused by the redundancy, self-similarity, and cluster structure inherent to those data sets. These phenomena manifest themselves in the data sets' representation in descriptor space, which is termed the data set topology. A methodology for the characterization of data set topology based on spatial statistics is introduced. The method is nonparametric and can deal with arbitrary distributions of descriptor values. With this methodology it is possible to associate differences in virtual screening performance on different data sets with differences in data set topology. Moreover, the better virtual screening performance of certain descriptors can be explained by their ability of representing the benchmark data sets by a more favorable topology. Finally it is shown, that the composition of some benchmark data sets causes topologies that lead to overoptimistic validation results even in very "simple" descriptor spaces. Spatial statistics analysis as proposed here facilitates the detection of such biased data sets and may provide a tool for the future design of unbiased benchmark data sets.

show abstract

Comparison of topological descriptors for similarity-based virtual screening using multiple bioactive reference structures

Cited by 245 publications

References 31 publications

Quantifying the Relationships among Drug Classes

Quantifying the Relationships among Drug Classes

Analysis and use of fragment-occurrence data in similarity-based virtual screening

Impact of Benchmark Data Set Topology on the Validation of Virtual Screening Methods: Exploration and Quantification by Spatial Statistics

Contact Info

Product

Resources

About