Charged aerosol detector response modeling for fatty acids based on experimental settings and molecular features: a machine learning approach
Authors
Pawellek, RubenKrmar, Jovana

Leistner, Adrian
Đajić, Nevena

Otašević, Biljana

Protić, Ana

Holzgrabe, Ulrike

Article (Published version)
Metadata
Show full item recordAbstract
The charged aerosol detector (CAD) is the latest representative of aerosol-based detectors that generate a response independent of the analytes’ chemical structure. This study was aimed at accurately predicting the CAD response of homologous fatty acids under varying experimental conditions. Fatty acids from C12 to C18 were used as model substances due to semivolatile characterics that caused non-uniform CAD behaviour. Considering both experimental conditions and molecular descriptors, a mixed quantitative structure–property relationship (QSPR) modeling was performed using Gradient Boosted Trees (GBT ). The ensemble of 10 decisions trees (learning rate set at 0.55, the maximal depth set at 5, and the sample rate set at 1.0) was able to explain approximately 99% (Q2: 0.987, RMSE: 0.051) of the observed variance in CAD responses. Validation using an external test compound confirmed the high predic- tive ability of the model established (R2: 0.990, RMSEP: 0.050). With respect to ...the intrinsic attribute selection strategy, GBT used almost all independent variables during model building. Finally, it attributed the highest importance to the power function value, the flow rate of the mobile phase, evaporation temperature, the content of the organic solvent in the mobile phase and the molecular descriptors such as molecular weight (MW ), Radial Distribution Func- tion—080/weighted by mass (RDF080m) and average coefficient of the last eigenvector from distance/detour matrix (Ve2_D/Dt). The identification of the factors most relevant to the CAD responsiveness has contributed to a better understanding of the underlying mechanisms of signal generation. An increased CAD response that was obtained for acetone as organic modifier demonstrated its potential to replace the more expensive and environmentally harmful acetonitrile.
Keywords:
Fatty acids / Charged aerosol detector (CAD) / Gradient boosted trees (GBT) / High-performance liquid chromatography (HPLC) / Quantitative structure–property relationship modeling (QSPR)Source:
Journal of Cheminformatics, 2021, 13, 1Publisher:
- BioMed Central Ltd
Funding / projects:
- DAAD PPP Program for Project-Related Personal Exchange with Serbia
- Fund of the University of Wuerzburg
- Ministry of Education, Science and Technological Development, Republic of Serbia, Grant no. 200161 (University of Belgrade, Faculty of Pharmacy) (RS-200161)
DOI: 10.1186/s13321-021-00532-0
ISSN: 1758-2946
WoS: 000673977300001
Scopus: 2-s2.0-85110487489
Collections
Institution/Community
PharmacyTY - JOUR AU - Pawellek, Ruben AU - Krmar, Jovana AU - Leistner, Adrian AU - Đajić, Nevena AU - Otašević, Biljana AU - Protić, Ana AU - Holzgrabe, Ulrike PY - 2021 UR - https://farfar.pharmacy.bg.ac.rs/handle/123456789/3926 AB - The charged aerosol detector (CAD) is the latest representative of aerosol-based detectors that generate a response independent of the analytes’ chemical structure. This study was aimed at accurately predicting the CAD response of homologous fatty acids under varying experimental conditions. Fatty acids from C12 to C18 were used as model substances due to semivolatile characterics that caused non-uniform CAD behaviour. Considering both experimental conditions and molecular descriptors, a mixed quantitative structure–property relationship (QSPR) modeling was performed using Gradient Boosted Trees (GBT ). The ensemble of 10 decisions trees (learning rate set at 0.55, the maximal depth set at 5, and the sample rate set at 1.0) was able to explain approximately 99% (Q2: 0.987, RMSE: 0.051) of the observed variance in CAD responses. Validation using an external test compound confirmed the high predic- tive ability of the model established (R2: 0.990, RMSEP: 0.050). With respect to the intrinsic attribute selection strategy, GBT used almost all independent variables during model building. Finally, it attributed the highest importance to the power function value, the flow rate of the mobile phase, evaporation temperature, the content of the organic solvent in the mobile phase and the molecular descriptors such as molecular weight (MW ), Radial Distribution Func- tion—080/weighted by mass (RDF080m) and average coefficient of the last eigenvector from distance/detour matrix (Ve2_D/Dt). The identification of the factors most relevant to the CAD responsiveness has contributed to a better understanding of the underlying mechanisms of signal generation. An increased CAD response that was obtained for acetone as organic modifier demonstrated its potential to replace the more expensive and environmentally harmful acetonitrile. PB - BioMed Central Ltd T2 - Journal of Cheminformatics T1 - Charged aerosol detector response modeling for fatty acids based on experimental settings and molecular features: a machine learning approach VL - 13 IS - 1 DO - 10.1186/s13321-021-00532-0 ER -
@article{ author = "Pawellek, Ruben and Krmar, Jovana and Leistner, Adrian and Đajić, Nevena and Otašević, Biljana and Protić, Ana and Holzgrabe, Ulrike", year = "2021", abstract = "The charged aerosol detector (CAD) is the latest representative of aerosol-based detectors that generate a response independent of the analytes’ chemical structure. This study was aimed at accurately predicting the CAD response of homologous fatty acids under varying experimental conditions. Fatty acids from C12 to C18 were used as model substances due to semivolatile characterics that caused non-uniform CAD behaviour. Considering both experimental conditions and molecular descriptors, a mixed quantitative structure–property relationship (QSPR) modeling was performed using Gradient Boosted Trees (GBT ). The ensemble of 10 decisions trees (learning rate set at 0.55, the maximal depth set at 5, and the sample rate set at 1.0) was able to explain approximately 99% (Q2: 0.987, RMSE: 0.051) of the observed variance in CAD responses. Validation using an external test compound confirmed the high predic- tive ability of the model established (R2: 0.990, RMSEP: 0.050). With respect to the intrinsic attribute selection strategy, GBT used almost all independent variables during model building. Finally, it attributed the highest importance to the power function value, the flow rate of the mobile phase, evaporation temperature, the content of the organic solvent in the mobile phase and the molecular descriptors such as molecular weight (MW ), Radial Distribution Func- tion—080/weighted by mass (RDF080m) and average coefficient of the last eigenvector from distance/detour matrix (Ve2_D/Dt). The identification of the factors most relevant to the CAD responsiveness has contributed to a better understanding of the underlying mechanisms of signal generation. An increased CAD response that was obtained for acetone as organic modifier demonstrated its potential to replace the more expensive and environmentally harmful acetonitrile.", publisher = "BioMed Central Ltd", journal = "Journal of Cheminformatics", title = "Charged aerosol detector response modeling for fatty acids based on experimental settings and molecular features: a machine learning approach", volume = "13", number = "1", doi = "10.1186/s13321-021-00532-0" }
Pawellek, R., Krmar, J., Leistner, A., Đajić, N., Otašević, B., Protić, A.,& Holzgrabe, U.. (2021). Charged aerosol detector response modeling for fatty acids based on experimental settings and molecular features: a machine learning approach. in Journal of Cheminformatics BioMed Central Ltd., 13(1). https://doi.org/10.1186/s13321-021-00532-0
Pawellek R, Krmar J, Leistner A, Đajić N, Otašević B, Protić A, Holzgrabe U. Charged aerosol detector response modeling for fatty acids based on experimental settings and molecular features: a machine learning approach. in Journal of Cheminformatics. 2021;13(1). doi:10.1186/s13321-021-00532-0 .
Pawellek, Ruben, Krmar, Jovana, Leistner, Adrian, Đajić, Nevena, Otašević, Biljana, Protić, Ana, Holzgrabe, Ulrike, "Charged aerosol detector response modeling for fatty acids based on experimental settings and molecular features: a machine learning approach" in Journal of Cheminformatics, 13, no. 1 (2021), https://doi.org/10.1186/s13321-021-00532-0 . .