Machine learning proteochemometric models for Cereblon glue activity predictions
Targeted protein degradation (TPD) is a rapidly developing drug discovery technique with unique efficacy and target scope stemming from its degradation-based activity. Molecular glue degraders are a promising arm of TPD, as evidenced by the FDA-approved therapeutics within this class, the increasing...
Saved in:
| Main Authors: | , , , , , , , , |
|---|---|
| Format: | Article |
| Language: | English |
| Published: |
Elsevier
2024-12-01
|
| Series: | Artificial Intelligence in the Life Sciences |
| Subjects: | |
| Online Access: | http://www.sciencedirect.com/science/article/pii/S2667318524000072 |
| Tags: |
Add Tag
No Tags, Be the first to tag this record!
|
| _version_ | 1846123259045609472 |
|---|---|
| author | Francis J. Prael, III Jiayi Cox Noé Sturm Peter Kutchukian William C. Forrester Gregory Michaud Jutta Blank Lingling Shen Raquel Rodríguez-Pérez |
| author_facet | Francis J. Prael, III Jiayi Cox Noé Sturm Peter Kutchukian William C. Forrester Gregory Michaud Jutta Blank Lingling Shen Raquel Rodríguez-Pérez |
| author_sort | Francis J. Prael, III |
| collection | DOAJ |
| description | Targeted protein degradation (TPD) is a rapidly developing drug discovery technique with unique efficacy and target scope stemming from its degradation-based activity. Molecular glue degraders are a promising arm of TPD, as evidenced by the FDA-approved therapeutics within this class, the increasing number of degraders in clinical development, and their predisposition to drug-likeness. Cereblon (CRBN) glue degraders mediate target degradation by generating a neomorphic interface between CRBN and a protein of interest. While promising, the complicated nature of this CRBN-glue-target ternary complex makes the rational design of molecular glue degraders challenging. For other drug modalities, predictive modeling has been established to leverage existing activity data and generate quantitative structure-activity relationships (QSAR). However, the applicability of QSAR strategies for glues remains under-investigated. Herein, machine learning methodologies were developed to predict glue-mediated recruitment of CRBN to target proteins and achieved promising performance. Generated models leveraged more than a hundred internal screening campaigns across thousands of CRBN glues to predict glue-mediated recruitment of targets to CRBN. Our results show that recruitment activity of CRBN glue degraders can be modeled by machine learning, with 89 % of models producing an area under the receiver operating characteristic curve (ROC AUC) > 0.8 and 70 % of models producing a Matthew's correlation coefficient (MCC) > 0.2 for these primary screening data. Importantly, our findings also indicate that the combination of compound and protein descriptors in the so-called proteochemometric models improves performance, with >80 % of the models exhibiting higher ROC AUC and MCC values than per-target models only based on compound information. Hence, our investigations suggest that proteochemometric modeling is a successful approach for molecular glue degraders. The proposed machine learning strategies can aid compound prioritization based on recruitment efficacy and target selectivity, thus have the potential to facilitate the design and discovery of therapeutic CRBN molecular glues. |
| format | Article |
| id | doaj-art-55a983e8b68c47b29a24a8f5a592f2c3 |
| institution | Kabale University |
| issn | 2667-3185 |
| language | English |
| publishDate | 2024-12-01 |
| publisher | Elsevier |
| record_format | Article |
| series | Artificial Intelligence in the Life Sciences |
| spelling | doaj-art-55a983e8b68c47b29a24a8f5a592f2c32024-12-14T06:33:53ZengElsevierArtificial Intelligence in the Life Sciences2667-31852024-12-016100100Machine learning proteochemometric models for Cereblon glue activity predictionsFrancis J. Prael, III0Jiayi Cox1Noé Sturm2Peter Kutchukian3William C. Forrester4Gregory Michaud5Jutta Blank6Lingling Shen7Raquel Rodríguez-Pérez8Novartis Biomedical Research, Novartis Campus, 02139 Cambridge MA, USANovartis Biomedical Research, Novartis Campus, 02139 Cambridge MA, USANovartis Biomedical Research, Novartis Campus, 4002 Basel, SwitzerlandNovartis Biomedical Research, Novartis Campus, 02139 Cambridge MA, USANovartis Biomedical Research, Novartis Campus, 02139 Cambridge MA, USANovartis Biomedical Research, Novartis Campus, 02139 Cambridge MA, USANovartis Biomedical Research, Novartis Campus, 02139 Cambridge MA, USANovartis Biomedical Research, Novartis Campus, 02139 Cambridge MA, USA; Corresponding authors.Novartis Biomedical Research, Novartis Campus, 4002 Basel, Switzerland; Corresponding authors.Targeted protein degradation (TPD) is a rapidly developing drug discovery technique with unique efficacy and target scope stemming from its degradation-based activity. Molecular glue degraders are a promising arm of TPD, as evidenced by the FDA-approved therapeutics within this class, the increasing number of degraders in clinical development, and their predisposition to drug-likeness. Cereblon (CRBN) glue degraders mediate target degradation by generating a neomorphic interface between CRBN and a protein of interest. While promising, the complicated nature of this CRBN-glue-target ternary complex makes the rational design of molecular glue degraders challenging. For other drug modalities, predictive modeling has been established to leverage existing activity data and generate quantitative structure-activity relationships (QSAR). However, the applicability of QSAR strategies for glues remains under-investigated. Herein, machine learning methodologies were developed to predict glue-mediated recruitment of CRBN to target proteins and achieved promising performance. Generated models leveraged more than a hundred internal screening campaigns across thousands of CRBN glues to predict glue-mediated recruitment of targets to CRBN. Our results show that recruitment activity of CRBN glue degraders can be modeled by machine learning, with 89 % of models producing an area under the receiver operating characteristic curve (ROC AUC) > 0.8 and 70 % of models producing a Matthew's correlation coefficient (MCC) > 0.2 for these primary screening data. Importantly, our findings also indicate that the combination of compound and protein descriptors in the so-called proteochemometric models improves performance, with >80 % of the models exhibiting higher ROC AUC and MCC values than per-target models only based on compound information. Hence, our investigations suggest that proteochemometric modeling is a successful approach for molecular glue degraders. The proposed machine learning strategies can aid compound prioritization based on recruitment efficacy and target selectivity, thus have the potential to facilitate the design and discovery of therapeutic CRBN molecular glues.http://www.sciencedirect.com/science/article/pii/S2667318524000072Machine learningGluesTargeted protein degradationCereblonProteochemometric modelsChemogenomics |
| spellingShingle | Francis J. Prael, III Jiayi Cox Noé Sturm Peter Kutchukian William C. Forrester Gregory Michaud Jutta Blank Lingling Shen Raquel Rodríguez-Pérez Machine learning proteochemometric models for Cereblon glue activity predictions Artificial Intelligence in the Life Sciences Machine learning Glues Targeted protein degradation Cereblon Proteochemometric models Chemogenomics |
| title | Machine learning proteochemometric models for Cereblon glue activity predictions |
| title_full | Machine learning proteochemometric models for Cereblon glue activity predictions |
| title_fullStr | Machine learning proteochemometric models for Cereblon glue activity predictions |
| title_full_unstemmed | Machine learning proteochemometric models for Cereblon glue activity predictions |
| title_short | Machine learning proteochemometric models for Cereblon glue activity predictions |
| title_sort | machine learning proteochemometric models for cereblon glue activity predictions |
| topic | Machine learning Glues Targeted protein degradation Cereblon Proteochemometric models Chemogenomics |
| url | http://www.sciencedirect.com/science/article/pii/S2667318524000072 |
| work_keys_str_mv | AT francisjpraeliii machinelearningproteochemometricmodelsforcereblonglueactivitypredictions AT jiayicox machinelearningproteochemometricmodelsforcereblonglueactivitypredictions AT noesturm machinelearningproteochemometricmodelsforcereblonglueactivitypredictions AT peterkutchukian machinelearningproteochemometricmodelsforcereblonglueactivitypredictions AT williamcforrester machinelearningproteochemometricmodelsforcereblonglueactivitypredictions AT gregorymichaud machinelearningproteochemometricmodelsforcereblonglueactivitypredictions AT juttablank machinelearningproteochemometricmodelsforcereblonglueactivitypredictions AT linglingshen machinelearningproteochemometricmodelsforcereblonglueactivitypredictions AT raquelrodriguezperez machinelearningproteochemometricmodelsforcereblonglueactivitypredictions |