Cabrera-Leyva L, Madera QJC, García-Jacas CR, Marrero-Ponce Y

Language: Spanish
References: 9
Page: 197-205
PDF size: 207.72 Kb.

ABSTRACT

The QSAR studies defined in the literature are based on uni-modal approaches and do not consider datasets with different chemical information. Thus, this research has as objective to apply and analyze the behavior of multi-modal approaches when QSAR studies are carried out. To this end, a compound dataset with hepatotoxicity activity was employed and four modalities were built considering molecular descriptors based on different mathematical theories. Also, several predictive models were developed taking into account both uni-modal and multi-modal approaches by using classification algorithms reported in the literature and implemented in R language. The parameters of these algorithms with the procedure "parameter tuning with repeated grid-search cross-validation" were optimized, while the strategy 10-fold cross-validation with 10 repetitions was used to corroborate the predictive accuracy of the models. As result of this study it can be stated that the behavior of the models based on multi-modal approach present significant differences with to those models developed from uni-modal approaches.

REFERENCES

1. Marrero Y, Santiago O-M, López Y-M, Barigye S-J, Torrens F. "Derivatives in discrete mathematics: a novel graph-theoretical invariant for generating new 2/3D molecular descriptors. I. Theory and QSPR application". Journal of computer-aided molecular design. vol. 26, pp. 1229-1246, 2012.

2. Ray B, Henaff M, Ma S, Efstathiadis E, Peskin E-R, Picone M, Poli T, Aliferis C-F, Statnikov A. "Information content and analysis methods for Multi-Modal High-Throughput Biomedical Data". Scientific reports. vol. 4, 2014.

3. Liew C-Y, Lim Y-C, Yap C-W. "Mixed learning algorithms and features ensemble in hepatotoxicity prediction". Journal of computer-aided molecular design. vol. 25, pp. 855-871, 2011.

4. Mauri A, Consonni V, Pavan M, Todeschini R. "Dragon software: An easy approach to molecular descriptor calculations". Match. vol. 56, pp. 237-248, 2006.

5. Yap CW. "PaDEL?descriptor: An open source software to calculate molecular descriptors and fingerprints". Journal of computational chemistry. vol. 32, pp. 1466-1474, 2011.

6. García C. R, Marrero Y, Ponce L, Barigye S. J, Valdés J. R, Contreras E. "QuBiLS?MIDAS: A parallel free?software for molecular descriptors computation based on multilinear algebraic maps". Journal of computational chemistry. vol. 35, pp. 1395-1409, 2014.

7. Krstajic D, Buturovic L. J, Leahy D. E, Thomas S. "Cross-validation pitfalls when selecting and assessing regression and classification models". Journal of cheminformatics. vol. 6, pp. 1-15, 2014.

8. Alcalá J, Fernández A, Luengo J, Derrac J, García S, Sánchez L, Herrera F. "Keel data-mining software tool: Data set repository, integration of algorithms and experimental analysis framework". Journal of Multiple-Valued Logic and Soft Computing. vol. 17, pp. 255-287, 2010.

9. Sheskin D. "Handbook of parametric and nonparametric statistical procedures, chapman & hall". presented at the CRC, 2003.