i would like to ask, wheather autocorrelation desciptors are sub type of 2d or 3d desciptors , they are offered by many platforms to calculate without demon about the differnce between them and padel and other desciptors
There are both 2D and 3D autocorrelation descriptors.
They are considered 2D if only the connections and not the position of the atoms in the 3-dimensional space is taken into account.
The most common 2D autocorrelation descriptors are:
- Moreau-Broto autocorrelation descriptors (Moreau, J. L., & Broto, P. (1980). Autocorrelation of molecular structures: Application to SAR studies. Nouveau Journal de Chimie, 4, 757–764)
- Moran autocorrelation descriptors (Moran, P. A. P. (1950). Notes on Continuous Stochastic Phenomena. Biometrika, 37(1/2), 17. https://doi.org/10.2307/2332142)
- Geary autocorrelation descriptors (Geary, R. C. (1954). The Contiguity Ratio and Statistical Mapping. The Incorporated Statistician, 5(3), 115. https://doi.org/10.2307/2986645)
An example of 3D autocorrelation descriptors is provided by Klein, C. T., Kaiser, D., & Ecker, G. (2004). Topological Distance Based 3D Descriptors for Use in QSAR and Diversity Analysis. Journal of Chemical Information and Computer Sciences, 44(1), 200–209. https://doi.org/10.1021/ci0256236.
You can check the list of molecular descriptors divided in logical block calculated by alvaDesc here: https://www.alvascience.com/alvadesc-descriptors/
AlvaDesc is provided with a detailed help including the description and references of all the calculated descriptors.
Chapter alvaDesc: A Tool to Calculate and Analyze Molecular Descript...
Chapter Molecular Descriptors
thank you so much for the answer .
May i ask about the competitive edge of autocorrelation desciptors over commonly used one like padel , Estate, Mopac,CDK, RDkit?
thanks in advance Andrea Mauri
Autocorrelation descriptors are a heterogeneous class of molecular descriptors but you should consider that many types of molecular descriptors have been proposed in the literature.
See:
Chapter Molecular Descriptors for Chemoinformatics
andBook Handbook of molecular descriptors. WileyVCH, Weinheim
Additionally autocorrelation descriptors are a class of molecular descriptors, while padel , Estate, Mopac,CDK, RDkit are software or toolkits that can calculate some molecular descriptors.
If I well remember some of the tools/toolkits you cited can calculate some autocorrelation descriptors.
- Badges
- Science topic