## Abstract Molecular similarity measures are important for many cheminformatics applications like ligandβbased virtual screening and quantitative structureβproperty relationships. Graph kernels are formal similarity measures defined directly on graphs, such as the (annotated) molecular structure g
Linear notations and molecular graph similarity
β Scribed by William C. Herndon; Steven H. Bertz
- Publisher
- John Wiley and Sons
- Year
- 1987
- Tongue
- English
- Weight
- 713 KB
- Volume
- 8
- Category
- Article
- ISSN
- 0192-8651
No coin nor oath required. For personal study only.
β¦ Synopsis
l b o simple linear notation systems are suggested to encode molecular structure including stereochemical elements. Both systems give rise to a unique numbering of the molecular graph, and thus also lead to a unique linear notation. Both linear notation systems are extremely compact and require only standard chemical symbols. A string comparison technique is developed to aeasure the similarity of two molecular linear notations. This procedure allows one to define a molecular similarity index with values that range from zero to unity, the zero value characterizing complete dissimilarity and the value of unity denoting identity. The notation and similarity index procedures are applied to several small molecular structures.
π SIMILAR VOLUMES
We outline a class of graph similarity measures that uses the structural similarity of local neighborhoods to derive pairwise similarity scores for the nodes of two different graphs, and present a related similarity measure that uses a linear update to generate both node and edge similarity scores.