Understanding single-protein fouling in micro- and ultrafiltration systems via machine-learning-based models
Protein fouling is a complex mechanism. To enhance understanding on protein fouling of membranes, the target of this study is twofold: (i) to determine the relative influences of parameters via the Random Forest (RF) model and (ii) to evaluate the predictive capability of the Neural Network (NN) mod...
| Main Authors: | , , |
|---|---|
| Other Authors: | |
| Format: | Journal Article |
| Language: | English |
| Published: |
2023
|
| Subjects: | |
| Online Access: | https://hdl.handle.net/10356/170058 |
| _version_ | 1824453920814006272 |
|---|---|
| author | Tanudjaja, Henry Jonathan Ng, Angie Qi Qi Chew, Jia Wei |
| author2 | School of Chemistry, Chemical Engineering and Biotechnology |
| author_facet | School of Chemistry, Chemical Engineering and Biotechnology Tanudjaja, Henry Jonathan Ng, Angie Qi Qi Chew, Jia Wei |
| author_sort | Tanudjaja, Henry Jonathan |
| collection | NTU |
| description | Protein fouling is a complex mechanism. To enhance understanding on protein fouling of membranes, the target of this study is twofold: (i) to determine the relative influences of parameters via the Random Forest (RF) model and (ii) to evaluate the predictive capability of the Neural Network (NN) model. Membrane pore size is the most dominant influence on fouling followed by transmembrane pressure (TMP), while membrane configuration (i.e., flat-sheet, hollow fiber, or tubular) is the least dominant. The NN model gives modest predictive capability despite inconsistencies and variabilities of the experimental setups and protocols, which invariably affects the important parameters in the database compiled from past publications. The database was divided into microfiltration (MF) and ultrafiltration (UF) subsets based on the membrane pore size values. It was found that the dominant parameters for permeate flux are different, with membrane pore size and protein concentration being dominant for MF and UF, respectively, while TMP is dominant for protein rejection for both cases. For permeate flux, membrane material is the most dominant parameter for the non-BSA database, while membrane pore size remains the most dominant parameter for protein rejection regardless of the protein used. Results show that such data-driven RF and NN models can enhance the understanding on the relative dominance of the parameters on different phenomena and provide adequate prediction of protein fouling, in the absence of any governing equations. |
| first_indexed | 2025-02-19T03:14:05Z |
| format | Journal Article |
| id | ntu-10356/170058 |
| institution | Nanyang Technological University |
| language | English |
| last_indexed | 2025-02-19T03:14:05Z |
| publishDate | 2023 |
| record_format | dspace |
| spelling | ntu-10356/1700582023-08-22T08:59:58Z Understanding single-protein fouling in micro- and ultrafiltration systems via machine-learning-based models Tanudjaja, Henry Jonathan Ng, Angie Qi Qi Chew, Jia Wei School of Chemistry, Chemical Engineering and Biotechnology Singapore Membrane Technology Centre Nanyang Environment and Water Research Institute Engineering::Chemical engineering Database Systems Forestry Protein fouling is a complex mechanism. To enhance understanding on protein fouling of membranes, the target of this study is twofold: (i) to determine the relative influences of parameters via the Random Forest (RF) model and (ii) to evaluate the predictive capability of the Neural Network (NN) model. Membrane pore size is the most dominant influence on fouling followed by transmembrane pressure (TMP), while membrane configuration (i.e., flat-sheet, hollow fiber, or tubular) is the least dominant. The NN model gives modest predictive capability despite inconsistencies and variabilities of the experimental setups and protocols, which invariably affects the important parameters in the database compiled from past publications. The database was divided into microfiltration (MF) and ultrafiltration (UF) subsets based on the membrane pore size values. It was found that the dominant parameters for permeate flux are different, with membrane pore size and protein concentration being dominant for MF and UF, respectively, while TMP is dominant for protein rejection for both cases. For permeate flux, membrane material is the most dominant parameter for the non-BSA database, while membrane pore size remains the most dominant parameter for protein rejection regardless of the protein used. Results show that such data-driven RF and NN models can enhance the understanding on the relative dominance of the parameters on different phenomena and provide adequate prediction of protein fouling, in the absence of any governing equations. Agency for Science, Technology and Research (A*STAR) Ministry of Education (MOE) We acknowledge funding from the A*STAR(Singapore) Advanced Manufacturing and Engineering(AME) under its Pharma Innovation Programme Singapore(PIPS) program(A20B3a0070), A*STAR(Singapore) Advanced Manufacturing and Engineering(AME) under its Individual Research Grant(IRG) program(A2083c0049),the Singapore Ministry of Education Academic Research Fund Tier 1 Grant(2019-T1-002-065;RG100/19), and the Singapore Ministry of Education Academic Research Fund Tier2 Grant(MOE-MOET2EP10120-0001). 2023-08-22T08:59:58Z 2023-08-22T08:59:58Z 2023 Journal Article Tanudjaja, H. J., Ng, A. Q. Q. & Chew, J. W. (2023). Understanding single-protein fouling in micro- and ultrafiltration systems via machine-learning-based models. Industrial and Engineering Chemistry Research, 62(19), 7610-7621. https://dx.doi.org/10.1021/acs.iecr.3c00275 0888-5885 https://hdl.handle.net/10356/170058 10.1021/acs.iecr.3c00275 2-s2.0-85160675636 19 62 7610 7621 en A20B3a0070 A2083c0049 2019-T1-002-065 RG100/19 MOE-MOET2EP10120-0001 Industrial and Engineering Chemistry Research © 2023 American Chemical Society. All rights reserved. |
| spellingShingle | Engineering::Chemical engineering Database Systems Forestry Tanudjaja, Henry Jonathan Ng, Angie Qi Qi Chew, Jia Wei Understanding single-protein fouling in micro- and ultrafiltration systems via machine-learning-based models |
| title | Understanding single-protein fouling in micro- and ultrafiltration systems via machine-learning-based models |
| title_full | Understanding single-protein fouling in micro- and ultrafiltration systems via machine-learning-based models |
| title_fullStr | Understanding single-protein fouling in micro- and ultrafiltration systems via machine-learning-based models |
| title_full_unstemmed | Understanding single-protein fouling in micro- and ultrafiltration systems via machine-learning-based models |
| title_short | Understanding single-protein fouling in micro- and ultrafiltration systems via machine-learning-based models |
| title_sort | understanding single protein fouling in micro and ultrafiltration systems via machine learning based models |
| topic | Engineering::Chemical engineering Database Systems Forestry |
| url | https://hdl.handle.net/10356/170058 |
| work_keys_str_mv | AT tanudjajahenryjonathan understandingsingleproteinfoulinginmicroandultrafiltrationsystemsviamachinelearningbasedmodels AT ngangieqiqi understandingsingleproteinfoulinginmicroandultrafiltrationsystemsviamachinelearningbasedmodels AT chewjiawei understandingsingleproteinfoulinginmicroandultrafiltrationsystemsviamachinelearningbasedmodels |