1996
DOI: 10.1007/bf01715525
|View full text |Cite
|
Sign up to set email alerts
|

Assessing the ability of chemical similarity measures to discriminate between active and inactive compounds

Abstract: A method for assessing the biological discriminating power of chemical similarity measures is presented. The main concern of this work was to develop an objective way of evaluating different similarity measures in terms of how well they distinguished between active and inactive compounds. In addition, we have explored the level of similarity required for optimal separation and commented on its implications for work in the field of chemical diversity studies. The results for one simple similarity measure showed… Show more

Help me understand this report

Search citation statements

Order By: Relevance

Paper Sections

Select...
1
1
1
1

Citation Types

3
41
0
2

Year Published

1996
1996
2017
2017

Publication Types

Select...
8
1

Relationship

0
9

Authors

Journals

citations
Cited by 50 publications
(46 citation statements)
references
References 3 publications
3
41
0
2
Order By: Relevance
“…TheT animoto coefficient is am easure of the similarity of different fingerprints, ranging from zero for no similarity to one for identical structures. [36] An ovel two-dimensional scaffold fingerprint (SFP) was presented by Rabal et al [37] for mining ring fragments (Figure 2). This method not only encodes the usual 2D and 3D descriptors,s uch as shape,t opology,h eteroatoms,b ridges, spirocyclic centers,d iversity points,s p 3 carbon atoms,a nd chirality,b ut also pharmacophoric features,s uch as the number of hydrogen-bond donors and acceptors and their relative orientation (vectors), as they play ac ritical role in drug discovery.S FP can be used 1) to identify alternative chemotypes,including bioisosteres,toareference ring either in avisual mode or by running quantitative similarity searches and 2) in chemotype-based diversity selection.…”
Section: Scaffolds and Scaffold Diversitymentioning
confidence: 98%
“…TheT animoto coefficient is am easure of the similarity of different fingerprints, ranging from zero for no similarity to one for identical structures. [36] An ovel two-dimensional scaffold fingerprint (SFP) was presented by Rabal et al [37] for mining ring fragments (Figure 2). This method not only encodes the usual 2D and 3D descriptors,s uch as shape,t opology,h eteroatoms,b ridges, spirocyclic centers,d iversity points,s p 3 carbon atoms,a nd chirality,b ut also pharmacophoric features,s uch as the number of hydrogen-bond donors and acceptors and their relative orientation (vectors), as they play ac ritical role in drug discovery.S FP can be used 1) to identify alternative chemotypes,including bioisosteres,toareference ring either in avisual mode or by running quantitative similarity searches and 2) in chemotype-based diversity selection.…”
Section: Scaffolds and Scaffold Diversitymentioning
confidence: 98%
“…If a set bit is considered as a feature present in the molecule, the Tanimoto coefficient is a measure of the number of common features in both molecules (31). A Tanimoto coefficient of >0.85 indicates that two molecules have similar activities (32).…”
Section: Methodsmentioning
confidence: 99%
“…Other studies have been carried out using high-throughput screening data sets to analyze the relationship between active and inactive compounds in light of their calculated similarity values [34,37,38]. These studies have illustrated that generally applicable similarity threshold values are not available as bioactivity markers and have highlighted the "0.85 myth".…”
Section: Intrinsic Caveatsmentioning
confidence: 97%