Untitled

Abstract

A new descriptor of molecular structure, EVA, for use in the derivation of robustly predictive QSAR relationships is described. It is based on theoretically derived normal coordinate frequencies, and has been used extensively and successfully in proprietary chemical discovery programmes within Shell Research. As a result of informal dissemination of the methodology, it is now being used successfully in related areas such as pharmaceutical drug discovery. Much of the experimental data used in development remain proprietary, and are not available for publication. This paper describes the method and illustrates its application to the calculation of nonproprietary data, log P(ow), in both explanatory and predictive modes. It will be followed by other publications illustrating its application to a range of data derived from biological systems.

The EVA spectral descriptor

“…EVA, and its associated data standardisation technique described below, was originally developed by workers at Shell Research Limited [18,19]. The rationale behind the use of such information as a molecular descriptor was "that a significant amount of information pertaining to molecular properties, in particular biological activity, might be contained within the molecular vibration wave-function, of which the vibrational spectrum is a fingerprint" [19].…”

“…The rationale behind the use of such information as a molecular descriptor was "that a significant amount of information pertaining to molecular properties, in particular biological activity, might be contained within the molecular vibration wave-function, of which the vibrational spectrum is a fingerprint" [19]. It is also the case that there is a close, albeit complex, relationship between molecular 3D structure and the corresponding IR spectrum, a characteristic that has made IR spectroscopy an extremely powerful tool for determining and identifying chemical structures.…”

“…There has thus been much interest in either tackling the alignment issue head-on [9,[11][12][13][14][15] or in seeking alternative molecular descriptors that are both sensitive to 3D-structure but that do not require structural superposition [11,13,14]. EVA [18][19][20][21][22][23] is one example of such a descriptor, based as it is upon molecular vibrations the characteristics of which are, in the absence of an external modifying influence such as a receptor, invariant to rotation and translation of the structures concerned. However, whilst EVA removes the need for superposition the method is sensitive to 3D structure although not to such an extent as a "true" 3D method such as CoMFA.…”

Untitled

“…The EVA descriptor [1,2] is derived from fundamental molecular vibrational frequencies of which there are 3N-6 (or 3N-5 for a linear compound such as acetylene) for an N-atom structure. The frequency values are projected onto a linear bounded frequency scale covering the range 1 to 4,000 cm -1 and then smeared out, and therefore overlapped, through the application of Gaussian kernels to each and every frequency value.…”

“…and an L of 5 cm -1 resulting in 800 (4,000/L) descriptor variables [1]. The number of variables is thus very much larger than the number of compounds in a standard QSAR dataset and the Partial least squares to Latent Structures (PLS) technique [5] has been used to provide a robust regression analysis.…”