GraphSig

GraphSig provides a novel significance model for graph mining.

• We extract neighborhood features from graphs and using a statistical background distribution, evaluate the significance of a subgraph.

• This technology has been used for the identification of active substructures of lead compounds and in the mining of protein-ligand complexes. This is illustrated above.

• It has also been used as a classifier in two situations: to classify molecules on the basis of their ADMET properties and to classify molecules into active and inactive. This is illustrated above.

This technology has been used in developing our tool, SigFinder.

References

1. Huahai He; Ambuj K. Singh; GraphRank: Statistical Modeling and Mining of Significant Subgraphs in the Feature Space. Proceedings of the 6th IEEE International Conference on Data Mining (ICDM), December, 2006, doi:10.1109/ICDM.2006.79
2. Sayan Ranu, Ambuj K. Singh; GraphSig: A Scalable Approach to Mining Significant Subgraphs in Large Graph Databases. , IEEE International Conference on Data Engineering, 2009, pp.844-855
3. Sayan Ranu; Ambuj Singh; Mining Statistically Significant Molecular Substructures for Efficient Molecular Classification., J. Chem. Inf. Model., 2009, 49(11), pp 2537–2550 DOI : 10.1021/ci900035z