Sie befinden Sich nicht im Netzwerk der Universität Paderborn. Der Zugriff auf elektronische Ressourcen ist gegebenenfalls nur via VPN oder Shibboleth (DFN-AAI) möglich. mehr Informationen...
MPCDDI: A Secure Multiparty Computation-Based Deep Learning Framework for Drug-Drug Interaction Predictions
Ist Teil von
Bioinformatics Research and Applications, 2023, Vol.13760, p.263-274
Ort / Verlag
Switzerland: Springer
Erscheinungsjahr
2023
Link zum Volltext
Quelle
Alma/SFX Local Collection
Beschreibungen/Notizen
Drug-drug interaction (DDI) is a key concern in drug development and pharmacovigilance. It is important to improve DDI predictions by integrating multi-source data from various pharmaceutical companies. Unfortunately, the data privacy and financial interest issues seriously influence the inter-institutional collaborations for DDI predictions. We propose MPCDDI, a secure multiparty computation-based deep learning framework for drug-drug interaction predictions. MPCDDI leverages the secret sharing technologies to incorporate the drug-related feature data from multiple institutions and develops a deep learning model for DDI predictions. In MPCDDI, all data transmission and deep learning operations are integrated into secure multiparty computation (MPC) frameworks to enable high-quality collaboration among pharmaceutical institutions without divulging private drug-related information. The results suggest that MPCDDI is superior to other five baselines and achieves the similar performance to that of the corresponding plaintext collaborations. More interestingly, MPCDDI significantly outperforms methods that use private data from the single institution. In summary, MPCDDI is an effective framework for promoting collaborative and privacy-preserving drug discovery.