Table of Contents
Fingerprint-Based And Graph-Based Methods For Similarity Searching In Chemical Databases
Overview Of Talk
Chemical Similarity Searching
Similarity Measures
Comparison Of Similarity Coefficients for 2D Similarity Searching
Similarity Coefficients For Fragment Bit-Strings
Do Different Coefficients Give Similar Rankings?
Ranking Of The Searches
Distance Matrix
Clustering The Resulting Distance Matrices
Results for Single Linkage
Results for Complete Linkage
Results for Group-Average
Overall Results: I
Overall Results: II
Data Fusion
Fusion Of Similarity Coefficients
Results Of Fused Searches
Which Coefficients Should One Use?
Conclusions
Applications Of Graph Matching In Databases Of 2D Chemical Structures
Types Of Isomorphism Procedure
Maximal Common Subgraph (MCS) Isomorphism Algorithms
Clique Detection
Maximum Clique: Largest Set Of Mutually Adjacent Vertices
Clique Detection Algorithms
RASCAL (Rapid Similarity Calculation)
Maximum Common Edge Subgraphs
Example MCS and MOS
Use Of Line-Graphs
Line Graph
Chemical Line Graph
K3 ?K1,3 Interchange
First-Level Upperbound Screening
Example Of Connectivity Distributions
Calculation Of |G12|
MCES-Based Similarity Measures
Second-Level Upperbound Screening
Use Of The Assignment Procedure
More Accurate Calculation Of |G12|
Initial Experiments
More Detailed Experiments
Conclusions
Acknowledgements
|