Designing Computational Models for Identification of Endocytic Protein Using Deep Learning
Supervisor: Dr. Farman Ali, Assistant Professor
Co-supervisor: Dr. Usman Hashmi, Senior Assistant Professor
Campus/School/Dept: BUIC E-8/BSEAS/CS
Supervisory Record:
- PhD Produced:01
- PhD Enrolled: Nill
- MS/MPhil Produced:02
- MS/MPhil Enrolled:02
Topic Brief Description:
Endocytic proteins play a vital role in cellular processes by mediating the internalization of molecules, particles, and fluids from the extracellular environment into the cell. This process, known as endocytosis, is crucial for maintaining cellular homeostasis, nutrient uptake, signal transduction, and immune responses. Endocytic proteins, such as clathrin, dynamin, and adaptins, are integral to the formation of vesicles that transport cargo into the cell. These proteins regulate vesicle budding, scission, and trafficking, ensuring efficient delivery of materials to appropriate intracellular compartments.
In the context of disease, endocytic proteins are of particular significance. Dysregulation of endocytosis has been implicated in various pathological conditions, including cancer, neurodegenerative diseases, and infections. For instance, many viruses exploit endocytic pathways for entry into host cells, highlighting the importance of these proteins in viral infections. Additionally, understanding the role of endocytic proteins in receptor-mediated endocytosis provides insights into drug delivery mechanisms and targeted therapies.
Research Objectives/Deliverables:
- Development of a novel computational model for accurate GP identification.
- Creation of two original primary sequence-based training and testing datasets.
- To evaluate and enhance the generalization ability of the model to ensure high prediction accuracy for diverse antigenic proteins.
Research Questions:
- How can we effectively combine feature engineering techniques and machine learning algorithms to improve the accuracy and specificity of endocytic protein prediction models?
- What are the key challenges and limitations of existing computational methods for endocytic protein identification, and how can we address these limitations to develop more robust and accurate models?
- How can we leverage large-scale protein sequence and functional annotation datasets to train and evaluate deep learning models for endocytic protein prediction?
Candidate’s Eligibility Profile:
- The applicant must have an MS/MPhil/Equivalent degree in MS (DS)/(CS) with CGPA >0.
- Experience with programming languages such as Fortran, C/C++, MATLAB, or Python is advantageous. Candidates should thrive in an international environment and have excellent communication skills to actively contribute to team research efforts.
- Proficiency in spoken and written English is essential. We value independence and responsibility while promoting teamwork and collaboration among colleagues.