The use of machine learning-based sequential virtual screening in the search of new ligands of 5-HT6 receptor
5-HT6 receptor takes part in learning and memory processes. For this reason, the use of ligands of this receptor in the treatment of neurodegenerative diseases such as Alzheimer's disease, depression or autism is being investigated. The development of machine learning (ML) and access to large c...
| Main Authors: | , , , |
|---|---|
| Format: | Article |
| Language: | Polish |
| Published: |
Polish Pharmaceutical Society
2023-04-01
|
| Series: | Farmacja Polska |
| Subjects: | |
| Online Access: | https://www.ptfarm.pl/download/?file=File%2FFarmacja+Polska%2F2022%2F11%2F01_OG_The_use_of_machine_learning.pdf |
| _version_ | 1797373850261913600 |
|---|---|
| author | Michał Sapa Alicja Gawalska Marcin Kołaczkowski Adam Bucki |
| author_facet | Michał Sapa Alicja Gawalska Marcin Kołaczkowski Adam Bucki |
| author_sort | Michał Sapa |
| collection | DOAJ |
| description | 5-HT6 receptor takes part in learning and memory processes. For this reason, the use of ligands of this receptor in the treatment of neurodegenerative diseases such as Alzheimer's disease, depression or autism is being investigated. The development of machine learning (ML) and access to large compound databases allow for the increasing use of these methods in search of new drugs. The use of ML in pre-clinical tests allows for a reduction in time and costs of drug discovery.
In this study, we used a sequential virtual screening approach in search of new structures with potential high affinity for the 5-HT6 receptor. Data from the ChEMBL database containing ligand binding affinities, measured as an inhibition constant (Ki), was used as the training dataset. Each step of the screening was based on machine learning models, the task of which was to classify compounds as potentially active and inactive. The first step included a ligand-based drug discovery (LBDD) approach, in which, using Klekota-Roth fingerprints and descriptors describing the chemical structure of the ligands, a classification model was developed to select a preliminary group of candidates from the Otava chemical compound database. In the second step, a structure-based drug discovery (SBDD) approach was used. For this purpose, compounds were docked to the homology model of the 5-HT6 receptor, developed using the AlphaFold algorithm and optimized by Induced-Fit Docking tool and molecular dynamics. Docking poses were scored by a trained Extra Trees classifier. Interactions of a reference ligand with 14 binding site residues were used as features for the trained model.
The use of machine learning as a scoring function allowed to improve the virtual screening parameters compared to the Glide GScore scoring function. Based on the obtained model, it was also confirmed that the location of a ligand near the Ser5.43 and Phe5.38 residues is important for binding the compound to the receptor. The procedure has allowed to select 20 candidates with new chemical structures compared to known ligands. In addition, the obtained compounds had a relatively low basic pKa compared to known ligands and thus may be suspected to have a low affinity for hERG channels and good brain penetration. |
| first_indexed | 2024-03-08T18:56:24Z |
| format | Article |
| id | doaj.art-f4f6b42daf114756a152b80dc4835d8b |
| institution | Directory Open Access Journal |
| issn | 0014-8261 |
| language | Polish |
| last_indexed | 2024-03-08T18:56:24Z |
| publishDate | 2023-04-01 |
| publisher | Polish Pharmaceutical Society |
| record_format | Article |
| series | Farmacja Polska |
| spelling | doaj.art-f4f6b42daf114756a152b80dc4835d8b2023-12-28T13:32:46ZpolPolish Pharmaceutical SocietyFarmacja Polska0014-82612023-04-01781160761410.32383/farmpol/161474161474The use of machine learning-based sequential virtual screening in the search of new ligands of 5-HT6 receptorMichał Sapa0https://orcid.org/0000-0001-9251-9261Alicja Gawalska1https://orcid.org/0000-0002-3131-9300Marcin Kołaczkowski2https://orcid.org/0000-0001-8402-1121Adam Bucki3https://orcid.org/0000-0003-0451-9814Katedra Chemii Farmaceutycznej, Uniwersytet Jagielloński - Collegium Medicum, Wydział Farmaceutyczny, PolskaKatedra Chemii Farmaceutycznej, Uniwersytet Jagielloński - Collegium Medicum, Wydział Farmaceutyczny, PolskaKatedra Chemii Farmaceutycznej, Uniwersytet Jagielloński - Collegium Medicum, Wydział Farmaceutyczny, PolskaKatedra Chemii Farmaceutycznej, Uniwersytet Jagielloński - Collegium Medicum, Wydział Farmaceutyczny, Polska5-HT6 receptor takes part in learning and memory processes. For this reason, the use of ligands of this receptor in the treatment of neurodegenerative diseases such as Alzheimer's disease, depression or autism is being investigated. The development of machine learning (ML) and access to large compound databases allow for the increasing use of these methods in search of new drugs. The use of ML in pre-clinical tests allows for a reduction in time and costs of drug discovery. In this study, we used a sequential virtual screening approach in search of new structures with potential high affinity for the 5-HT6 receptor. Data from the ChEMBL database containing ligand binding affinities, measured as an inhibition constant (Ki), was used as the training dataset. Each step of the screening was based on machine learning models, the task of which was to classify compounds as potentially active and inactive. The first step included a ligand-based drug discovery (LBDD) approach, in which, using Klekota-Roth fingerprints and descriptors describing the chemical structure of the ligands, a classification model was developed to select a preliminary group of candidates from the Otava chemical compound database. In the second step, a structure-based drug discovery (SBDD) approach was used. For this purpose, compounds were docked to the homology model of the 5-HT6 receptor, developed using the AlphaFold algorithm and optimized by Induced-Fit Docking tool and molecular dynamics. Docking poses were scored by a trained Extra Trees classifier. Interactions of a reference ligand with 14 binding site residues were used as features for the trained model. The use of machine learning as a scoring function allowed to improve the virtual screening parameters compared to the Glide GScore scoring function. Based on the obtained model, it was also confirmed that the location of a ligand near the Ser5.43 and Phe5.38 residues is important for binding the compound to the receptor. The procedure has allowed to select 20 candidates with new chemical structures compared to known ligands. In addition, the obtained compounds had a relatively low basic pKa compared to known ligands and thus may be suspected to have a low affinity for hERG channels and good brain penetration.https://www.ptfarm.pl/download/?file=File%2FFarmacja+Polska%2F2022%2F11%2F01_OG_The_use_of_machine_learning.pdfmolecular dockingmachine learningstructure-activity relationshipserotonin 6 receptor |
| spellingShingle | Michał Sapa Alicja Gawalska Marcin Kołaczkowski Adam Bucki The use of machine learning-based sequential virtual screening in the search of new ligands of 5-HT6 receptor Farmacja Polska molecular docking machine learning structure-activity relationship serotonin 6 receptor |
| title | The use of machine learning-based sequential virtual screening in the search of new ligands of 5-HT6 receptor |
| title_full | The use of machine learning-based sequential virtual screening in the search of new ligands of 5-HT6 receptor |
| title_fullStr | The use of machine learning-based sequential virtual screening in the search of new ligands of 5-HT6 receptor |
| title_full_unstemmed | The use of machine learning-based sequential virtual screening in the search of new ligands of 5-HT6 receptor |
| title_short | The use of machine learning-based sequential virtual screening in the search of new ligands of 5-HT6 receptor |
| title_sort | use of machine learning based sequential virtual screening in the search of new ligands of 5 ht6 receptor |
| topic | molecular docking machine learning structure-activity relationship serotonin 6 receptor |
| url | https://www.ptfarm.pl/download/?file=File%2FFarmacja+Polska%2F2022%2F11%2F01_OG_The_use_of_machine_learning.pdf |
| work_keys_str_mv | AT michałsapa theuseofmachinelearningbasedsequentialvirtualscreeninginthesearchofnewligandsof5ht6receptor AT alicjagawalska theuseofmachinelearningbasedsequentialvirtualscreeninginthesearchofnewligandsof5ht6receptor AT marcinkołaczkowski theuseofmachinelearningbasedsequentialvirtualscreeninginthesearchofnewligandsof5ht6receptor AT adambucki theuseofmachinelearningbasedsequentialvirtualscreeninginthesearchofnewligandsof5ht6receptor AT michałsapa useofmachinelearningbasedsequentialvirtualscreeninginthesearchofnewligandsof5ht6receptor AT alicjagawalska useofmachinelearningbasedsequentialvirtualscreeninginthesearchofnewligandsof5ht6receptor AT marcinkołaczkowski useofmachinelearningbasedsequentialvirtualscreeninginthesearchofnewligandsof5ht6receptor AT adambucki useofmachinelearningbasedsequentialvirtualscreeninginthesearchofnewligandsof5ht6receptor |