Chromatographic and computational assessment of lipophilicity using sum of ranking differences and generalized pair-correlation
Abstract
Lipophilicity (logP) represents one of the most studied and most frequently used fundamental physicochemical properties. At present there are several possibilities for its quantitative expression and many of them stems from chromatographic experiments. Numerous attempts have been made to compare different computational methods, chromatographic methods vs. computational approaches, as well as chromatographic methods and direct shake-flask procedure without definite results or these findings are not accepted generally. In the present work numerous chromatographically derived lipophilicity measures in combination with diverse computational methods were ranked and clustered using the novel variable discrimination and ranking approaches based on the sum of ranking differences and the generalized pair correlation method. Available literature logP data measured on HILIC, and classical reversed-phase combining different classes of compounds have been compared with most frequently used multivar...iate data analysis techniques (principal component and hierarchical cluster analysis) as well as with the conclusions in the original sources. Chromatographic lipophilicity measures obtained under typical reversed-phase conditions outperform the majority of computationally estimated logPs. Oppositely, in the case of HILIC none of the many proposed chromatographic indices overcomes any of the computationally assessed logPs. Only two of them (logk(min) and k(mm)) may be selected as recommended chromatographic lipophilicity measures. Both ranking approaches, sum of ranking differences and generalized pair correlation method, although based on different backgrounds, provides highly similar variable ordering and grouping leading to the same conclusions.
Keywords:
Lipophilicity / Multivariate data analysis / Sum of ranking differences / Generalized pair correlation method / High-performance liquid chromatographySource:
Journal of Chromatography A, 2015, 1380, 130-138Publisher:
- Elsevier Science Bv, Amsterdam
Funding / projects:
- Structure-properties relationships of natural and synthetic molecules and their metal complexes (RS-MESTD-Basic Research (BR or ON)-172017)
- OTKA [K112547]
Note:
- Supplementary material: http://cherry.chem.bg.ac.rs/handle/123456789/3354
DOI: 10.1016/j.chroma.2014.12.073
ISSN: 0021-9673
PubMed: 25595531
WoS: 000348893100016
Scopus: 2-s2.0-84921306440
Collections
Institution/Community
Hemijski fakultet / Faculty of ChemistryTY - JOUR AU - Andrić, Filip AU - Héberger, Karoly PY - 2015 UR - https://cherry.chem.bg.ac.rs/handle/123456789/1650 AB - Lipophilicity (logP) represents one of the most studied and most frequently used fundamental physicochemical properties. At present there are several possibilities for its quantitative expression and many of them stems from chromatographic experiments. Numerous attempts have been made to compare different computational methods, chromatographic methods vs. computational approaches, as well as chromatographic methods and direct shake-flask procedure without definite results or these findings are not accepted generally. In the present work numerous chromatographically derived lipophilicity measures in combination with diverse computational methods were ranked and clustered using the novel variable discrimination and ranking approaches based on the sum of ranking differences and the generalized pair correlation method. Available literature logP data measured on HILIC, and classical reversed-phase combining different classes of compounds have been compared with most frequently used multivariate data analysis techniques (principal component and hierarchical cluster analysis) as well as with the conclusions in the original sources. Chromatographic lipophilicity measures obtained under typical reversed-phase conditions outperform the majority of computationally estimated logPs. Oppositely, in the case of HILIC none of the many proposed chromatographic indices overcomes any of the computationally assessed logPs. Only two of them (logk(min) and k(mm)) may be selected as recommended chromatographic lipophilicity measures. Both ranking approaches, sum of ranking differences and generalized pair correlation method, although based on different backgrounds, provides highly similar variable ordering and grouping leading to the same conclusions. PB - Elsevier Science Bv, Amsterdam T2 - Journal of Chromatography A T1 - Chromatographic and computational assessment of lipophilicity using sum of ranking differences and generalized pair-correlation VL - 1380 SP - 130 EP - 138 DO - 10.1016/j.chroma.2014.12.073 ER -
@article{ author = "Andrić, Filip and Héberger, Karoly", year = "2015", abstract = "Lipophilicity (logP) represents one of the most studied and most frequently used fundamental physicochemical properties. At present there are several possibilities for its quantitative expression and many of them stems from chromatographic experiments. Numerous attempts have been made to compare different computational methods, chromatographic methods vs. computational approaches, as well as chromatographic methods and direct shake-flask procedure without definite results or these findings are not accepted generally. In the present work numerous chromatographically derived lipophilicity measures in combination with diverse computational methods were ranked and clustered using the novel variable discrimination and ranking approaches based on the sum of ranking differences and the generalized pair correlation method. Available literature logP data measured on HILIC, and classical reversed-phase combining different classes of compounds have been compared with most frequently used multivariate data analysis techniques (principal component and hierarchical cluster analysis) as well as with the conclusions in the original sources. Chromatographic lipophilicity measures obtained under typical reversed-phase conditions outperform the majority of computationally estimated logPs. Oppositely, in the case of HILIC none of the many proposed chromatographic indices overcomes any of the computationally assessed logPs. Only two of them (logk(min) and k(mm)) may be selected as recommended chromatographic lipophilicity measures. Both ranking approaches, sum of ranking differences and generalized pair correlation method, although based on different backgrounds, provides highly similar variable ordering and grouping leading to the same conclusions.", publisher = "Elsevier Science Bv, Amsterdam", journal = "Journal of Chromatography A", title = "Chromatographic and computational assessment of lipophilicity using sum of ranking differences and generalized pair-correlation", volume = "1380", pages = "130-138", doi = "10.1016/j.chroma.2014.12.073" }
Andrić, F.,& Héberger, K.. (2015). Chromatographic and computational assessment of lipophilicity using sum of ranking differences and generalized pair-correlation. in Journal of Chromatography A Elsevier Science Bv, Amsterdam., 1380, 130-138. https://doi.org/10.1016/j.chroma.2014.12.073
Andrić F, Héberger K. Chromatographic and computational assessment of lipophilicity using sum of ranking differences and generalized pair-correlation. in Journal of Chromatography A. 2015;1380:130-138. doi:10.1016/j.chroma.2014.12.073 .
Andrić, Filip, Héberger, Karoly, "Chromatographic and computational assessment of lipophilicity using sum of ranking differences and generalized pair-correlation" in Journal of Chromatography A, 1380 (2015):130-138, https://doi.org/10.1016/j.chroma.2014.12.073 . .