Identifying natural inhibitors against FUS protein in dementia through machine learning, molecular docking, and dynamics simulation
Dementia, a complex and debilitating spectrum of neurodegenerative diseases, presents a profound challenge in the quest for effective treatments. The FUS protein is well at the center of this problem, as it is frequently dysregulated in the various disorders. We chose a route of computational work t...
Saved in:
| Main Author: | |
|---|---|
| Format: | Article |
| Language: | English |
| Published: |
Frontiers Media S.A.
2025-02-01
|
| Series: | Frontiers in Neuroinformatics |
| Subjects: | |
| Online Access: | https://www.frontiersin.org/articles/10.3389/fninf.2024.1439090/full |
| Tags: |
Add Tag
No Tags, Be the first to tag this record!
|
| _version_ | 1832539936118538240 |
|---|---|
| author | Darwin Li |
| author_facet | Darwin Li |
| author_sort | Darwin Li |
| collection | DOAJ |
| description | Dementia, a complex and debilitating spectrum of neurodegenerative diseases, presents a profound challenge in the quest for effective treatments. The FUS protein is well at the center of this problem, as it is frequently dysregulated in the various disorders. We chose a route of computational work that involves targeting natural inhibitors of the FUS protein, offering a novel treatment strategy. We first reviewed the FUS protein's framework; early forecasting models using the AlphaFold2 and SwissModel algorithms indicated a loop-rich protein—a structure component correlating with flexibility. However, these models showed limitations, as reflected by inadequate ERRAT and Verify3D scores. Seeking enhanced accuracy, we turned to the I-TASSER suite, which delivered a refined structural model affirmed by robust validation metrics. With a reliable model in hand, our study utilized machine learning techniques, particularly the Random Forest algorithm, to navigate through a vast dataset of phytochemicals. This led to the identification of nimbinin, dehydroxymethylflazine, and several other compounds as potential FUS inhibitors. Notably, dehydroxymethylflazine and cleroindicin C identified during molecular docking analyses—facilitated by AutoDock Vina—for their high binding affinities and stability in interaction with the FUS protein, as corroborated by extensive molecular dynamics simulations. Originating from medicinal plants, these compounds are not only structurally compatible with the target protein but also adhere to pharmacokinetic profiles suitable for drug development, including optimal molecular weight and LogP values conducive to blood-brain barrier penetration. This computational exploration paves the way for subsequent experimental validation and highlights the potential of these natural compounds as innovative agents in the treatment of dementia. |
| format | Article |
| id | doaj-art-e45ae8c19dac4e258d293fec38df088a |
| institution | Kabale University |
| issn | 1662-5196 |
| language | English |
| publishDate | 2025-02-01 |
| publisher | Frontiers Media S.A. |
| record_format | Article |
| series | Frontiers in Neuroinformatics |
| spelling | doaj-art-e45ae8c19dac4e258d293fec38df088a2025-02-05T07:32:52ZengFrontiers Media S.A.Frontiers in Neuroinformatics1662-51962025-02-011810.3389/fninf.2024.14390901439090Identifying natural inhibitors against FUS protein in dementia through machine learning, molecular docking, and dynamics simulationDarwin LiDementia, a complex and debilitating spectrum of neurodegenerative diseases, presents a profound challenge in the quest for effective treatments. The FUS protein is well at the center of this problem, as it is frequently dysregulated in the various disorders. We chose a route of computational work that involves targeting natural inhibitors of the FUS protein, offering a novel treatment strategy. We first reviewed the FUS protein's framework; early forecasting models using the AlphaFold2 and SwissModel algorithms indicated a loop-rich protein—a structure component correlating with flexibility. However, these models showed limitations, as reflected by inadequate ERRAT and Verify3D scores. Seeking enhanced accuracy, we turned to the I-TASSER suite, which delivered a refined structural model affirmed by robust validation metrics. With a reliable model in hand, our study utilized machine learning techniques, particularly the Random Forest algorithm, to navigate through a vast dataset of phytochemicals. This led to the identification of nimbinin, dehydroxymethylflazine, and several other compounds as potential FUS inhibitors. Notably, dehydroxymethylflazine and cleroindicin C identified during molecular docking analyses—facilitated by AutoDock Vina—for their high binding affinities and stability in interaction with the FUS protein, as corroborated by extensive molecular dynamics simulations. Originating from medicinal plants, these compounds are not only structurally compatible with the target protein but also adhere to pharmacokinetic profiles suitable for drug development, including optimal molecular weight and LogP values conducive to blood-brain barrier penetration. This computational exploration paves the way for subsequent experimental validation and highlights the potential of these natural compounds as innovative agents in the treatment of dementia.https://www.frontiersin.org/articles/10.3389/fninf.2024.1439090/fulldementiaFUS proteinmachine learningmolecular dockingmolecular dynamic simulation |
| spellingShingle | Darwin Li Identifying natural inhibitors against FUS protein in dementia through machine learning, molecular docking, and dynamics simulation Frontiers in Neuroinformatics dementia FUS protein machine learning molecular docking molecular dynamic simulation |
| title | Identifying natural inhibitors against FUS protein in dementia through machine learning, molecular docking, and dynamics simulation |
| title_full | Identifying natural inhibitors against FUS protein in dementia through machine learning, molecular docking, and dynamics simulation |
| title_fullStr | Identifying natural inhibitors against FUS protein in dementia through machine learning, molecular docking, and dynamics simulation |
| title_full_unstemmed | Identifying natural inhibitors against FUS protein in dementia through machine learning, molecular docking, and dynamics simulation |
| title_short | Identifying natural inhibitors against FUS protein in dementia through machine learning, molecular docking, and dynamics simulation |
| title_sort | identifying natural inhibitors against fus protein in dementia through machine learning molecular docking and dynamics simulation |
| topic | dementia FUS protein machine learning molecular docking molecular dynamic simulation |
| url | https://www.frontiersin.org/articles/10.3389/fninf.2024.1439090/full |
| work_keys_str_mv | AT darwinli identifyingnaturalinhibitorsagainstfusproteinindementiathroughmachinelearningmoleculardockinganddynamicssimulation |