Loading...
| Friend's email | |
| Your name | |
| Your email | |
| enter code | |
This page was sent successfuly
- Type of Document: M.Sc. Thesis
- Language: Farsi
- Document No: 57758 (19)
- University: Sharif University of Technology
- Department: Computer Engineering
- Advisor(s): Hossein Khalaj, Babak
- Abstract:
- Selective molecule design is one of the great challenges in drug design. Recently, with the introduction of diffusion models, good progress has been made in discovering new pharmaceutical compounds. These tools can produce new and diverse compounds with desired biological properties and activities. In this research, we propose a new framework for designing a molecular linker, the goal of which is to connect two or more molecules with a linker that increases their binding affinity and selectivity. We use a graph neural network to encode molecular features and a diffusion model to generate linker candidates. Also, in this research, we apply the structural features of the protein pocket in graph encoding
- Keywords:
- Diffusion Model ; Binding Affinity ; Selectivity ; Molecular Linker Design ; Molecule Design
Digital Object List
محتواي کتاب
Bookmark