FLAP 4-Point Pharmacophore Fingerprints from GRID

FLAP 4-Point Pharmacophore Fingerprints from GRID 

Francesca Ferruccio, Jonathan S. Mason, Simone Sciabola, and Massimo Baroni Abstract 

Molecular similarity and diversity methods typically represent molecules by a vector of real-valued properties (molecular weight, logP etc.) or binary values (0 for absence, 1 for presence of a substructure feature for example) in a bit-string or 

Using 4-point pharmacophores enables chirality to be handled and adds some elements of volume/shape linked to electronic properties, increasing separation in similarity and diversity studies. There is a large increase in the number of potential pharmacophores that need to be considered. For example, using six possible feature types for each point, and 10 distance ranges (bins) for each feature-feature distance, the number of potential pharmacophores increases from 33 000 for 3-point pharmacophores to 9.7 million for 4-point pharmacophores [14—16]. Reducing the number of distance bins to seven reduces these numbers to 9000 and 2.3 million respectively. 

The granularity of conformational sampling, generally performed by torsional sam- 

---

I 4 FLAP 4-Point Pharmacophore Fingerprints from GRID 4.1... 

Perruccio, F., Mason, J.S., Sciabola, S. and Baroni, M. (2006) FLAP 4-point pharmacophore fingerprints from GRID, in Molecular Interaction Fields, Vol. 27 (ed. G. Crudani), Wiley-VCH Verlag GmbH, Weinheim, Germany. 

---