Machine Learning Scoring Functions for Drug Discovery from Experimental and Computer-Generated Protein–Ligand Structures: Towards Per-Target Scoring Functions
In recent years, machine learning has been proposed as a promising strategy to build accurate scoring functions for computational docking finalized to numerically empowered drug discovery. However, the latest studies have suggested that over-optimistic results had been reported due to the correlatio...
| Main Authors: | , , , |
|---|---|
| Format: | Article |
| Language: | English |
| Published: |
MDPI AG
2023-02-01
|
| Series: | Molecules |
| Subjects: | |
| Online Access: | https://www.mdpi.com/1420-3049/28/4/1661 |
| _version_ | 1797619015112196096 |
|---|---|
| author | Francesco Pellicani Diego Dal Ben Andrea Perali Sebastiano Pilati |
| author_facet | Francesco Pellicani Diego Dal Ben Andrea Perali Sebastiano Pilati |
| author_sort | Francesco Pellicani |
| collection | DOAJ |
| description | In recent years, machine learning has been proposed as a promising strategy to build accurate scoring functions for computational docking finalized to numerically empowered drug discovery. However, the latest studies have suggested that over-optimistic results had been reported due to the correlations present in the experimental databases used for training and testing. Here, we investigate the performance of an artificial neural network in binding affinity predictions, comparing results obtained using both experimental protein–ligand structures as well as larger sets of computer-generated structures created using commercial software. Interestingly, similar performances are obtained on both databases. We find a noticeable performance suppression when moving from random horizontal tests to vertical tests performed on target proteins not included in the training data. The possibility to train the network on relatively easily created computer-generated databases leads us to explore per-target scoring functions, trained and tested ad-hoc on complexes including only one target protein. Encouraging results are obtained, depending on the type of protein being addressed. |
| first_indexed | 2024-03-11T08:22:03Z |
| format | Article |
| id | doaj.art-555bfdce9c09412791542ff39606fa73 |
| institution | Directory Open Access Journal |
| issn | 1420-3049 |
| language | English |
| last_indexed | 2024-03-11T08:22:03Z |
| publishDate | 2023-02-01 |
| publisher | MDPI AG |
| record_format | Article |
| series | Molecules |
| spelling | doaj.art-555bfdce9c09412791542ff39606fa732023-11-16T22:21:30ZengMDPI AGMolecules1420-30492023-02-01284166110.3390/molecules28041661Machine Learning Scoring Functions for Drug Discovery from Experimental and Computer-Generated Protein–Ligand Structures: Towards Per-Target Scoring FunctionsFrancesco Pellicani0Diego Dal Ben1Andrea Perali2Sebastiano Pilati3Physics Division, School of Science and Technology, University of Camerino, I-62032 Camerino, MC, ItalyMedicinal Chemistry Unit, School of Pharmacy, University of Camerino, I-62032 Camerino, MC, ItalyPhysics Unit, School of Pharmacy, University of Camerino, I-62032 Camerino, MC, ItalyPhysics Division, School of Science and Technology, University of Camerino, I-62032 Camerino, MC, ItalyIn recent years, machine learning has been proposed as a promising strategy to build accurate scoring functions for computational docking finalized to numerically empowered drug discovery. However, the latest studies have suggested that over-optimistic results had been reported due to the correlations present in the experimental databases used for training and testing. Here, we investigate the performance of an artificial neural network in binding affinity predictions, comparing results obtained using both experimental protein–ligand structures as well as larger sets of computer-generated structures created using commercial software. Interestingly, similar performances are obtained on both databases. We find a noticeable performance suppression when moving from random horizontal tests to vertical tests performed on target proteins not included in the training data. The possibility to train the network on relatively easily created computer-generated databases leads us to explore per-target scoring functions, trained and tested ad-hoc on complexes including only one target protein. Encouraging results are obtained, depending on the type of protein being addressed.https://www.mdpi.com/1420-3049/28/4/1661molecular dockingscoring functionsmachine learning |
| spellingShingle | Francesco Pellicani Diego Dal Ben Andrea Perali Sebastiano Pilati Machine Learning Scoring Functions for Drug Discovery from Experimental and Computer-Generated Protein–Ligand Structures: Towards Per-Target Scoring Functions Molecules molecular docking scoring functions machine learning |
| title | Machine Learning Scoring Functions for Drug Discovery from Experimental and Computer-Generated Protein–Ligand Structures: Towards Per-Target Scoring Functions |
| title_full | Machine Learning Scoring Functions for Drug Discovery from Experimental and Computer-Generated Protein–Ligand Structures: Towards Per-Target Scoring Functions |
| title_fullStr | Machine Learning Scoring Functions for Drug Discovery from Experimental and Computer-Generated Protein–Ligand Structures: Towards Per-Target Scoring Functions |
| title_full_unstemmed | Machine Learning Scoring Functions for Drug Discovery from Experimental and Computer-Generated Protein–Ligand Structures: Towards Per-Target Scoring Functions |
| title_short | Machine Learning Scoring Functions for Drug Discovery from Experimental and Computer-Generated Protein–Ligand Structures: Towards Per-Target Scoring Functions |
| title_sort | machine learning scoring functions for drug discovery from experimental and computer generated protein ligand structures towards per target scoring functions |
| topic | molecular docking scoring functions machine learning |
| url | https://www.mdpi.com/1420-3049/28/4/1661 |
| work_keys_str_mv | AT francescopellicani machinelearningscoringfunctionsfordrugdiscoveryfromexperimentalandcomputergeneratedproteinligandstructurestowardspertargetscoringfunctions AT diegodalben machinelearningscoringfunctionsfordrugdiscoveryfromexperimentalandcomputergeneratedproteinligandstructurestowardspertargetscoringfunctions AT andreaperali machinelearningscoringfunctionsfordrugdiscoveryfromexperimentalandcomputergeneratedproteinligandstructurestowardspertargetscoringfunctions AT sebastianopilati machinelearningscoringfunctionsfordrugdiscoveryfromexperimentalandcomputergeneratedproteinligandstructurestowardspertargetscoringfunctions |