Don't miss our weekly PhD newsletter | Sign up now Don't miss our weekly PhD newsletter | Sign up now

  Deep learning enabled alternative drug design for cancer patients

   Faculty of Engineering, Computing and the Environment

This project is no longer listed on and may not be available.

Click here to search for PhD studentship opportunities
  Dr Farzana Rahman  Applications accepted all year round  Self-Funded PhD Students Only

About the Project

We invite applications from research students interested in learning and developing an understanding of deep learning-based solutions for drug design for cancer chemotherapy. Previous knowledge of biology or medical science is not mandatory. 

In this project, the PhD candidate will learn, develop, and apply deep-learning-based drug discovery techniques to find suitable alternatives for a popular drug formula (Taxol). They will then apply Natural Language Processing (NLP) techniques to find support for machine-recommended drug design. As part of this task, the candidate will investigate available methods and create a novel method to "interpret a large body of text at a pace". In addition, they will learn about keyword-directed search to perform topic modelling using named entity recognition and extended classification mechanisms. 

 The high impact of cancer on human life does not need any introduction. While chemical therapy (or chemotherapy) is a widely used practice of getting rid of cancerous cells, the side effects of the drugs used can be life changing. Chemotherapy is often used to prolong life or reduce symptoms. In a nutshell, a combination of chemical drugs circulates throughout the body of a patient and terminate any/all cancerous cell it may find. However, the limited options available for chemical combinations of drugs often face challenges in treating a diverse range of cancers in an even more diverse range of patient profiles. This project will use computational techniques, tools, and novel methods to prepare a protocol for providing alternative combinations and formulae for drugs.

 We have identified a safe drug, Taxol (paclitaxel), that has been approved by regulatory bodies. This drug, which is a natural plant alkaloid produced by yew trees, is considered one of the most effective chemotherapeutic drugs ever developed. Thus, it is widely used for the systemic treatment of various cancers (ovarian, breast, lung etc.). However, its very low solubility, along with a complex delivery system, induces inevitable severe side effects, which often relate to generalised toxicity in the patient body. Thus, finding an alternative formulation system delivery system that explicitly targets cancerous cells is a worthy scientific endeavour.

 The PhD Project

 As much investment has gone into cancer research, a wealth of literature is available. Using natural language processing techniques, the PhD candidate will analyse current literature on Taxol side effects and formulae. They will perform a similar analysis to find identical side effects from other drugs whose side effects were eventually reduced. Finally, the student will design, build, and train models to propose alternative usage of Taxol in ways that significantly reduce severe side effects while preserving bioactivities. 

 As a common criticism of a deep transfer model is its data-hungry nature and inherent non-interoperability, this project will seek to improve the interpretability and interoperability of the developed models.

 The candidate will design and use novel algorithms to identify and characterise polypeptides that specifically recognise cancer-specific cellular receptors. The study will then use machine learning models to recommend the best ways to link the polypeptide to Taxol.

 Using data produced by the collaborating laboratories that will conduct experiments in vitro and in vivo systems, the candidate will be able to determine the stability of the new taxol-polypeptide formula in different combinations of physiological factors. The outcome of this project will provide a promising functional formula for Taxol for targeted chemotherapy applications.

Student Eligibility

Applicants should have at least an Honours Degree at 2.1 or above (or equivalent) in a STEM discipline with interest in learning Artificial Intelligence/Data Science areas. In addition, the applicant must have previous programming experience in Python/R in an academic or industry setup. Prior knowledge or experience in Oncology is not a requirement. However, students without physiology or life-sciences backgrounds will be required to take some learning courses in relevant fields.

The project team:

This project will follow a collaboration between multiple collaborating labs in Europe, Asia, and the United States of America.

Key partners for this project are Prof. Denis Murphy (University of South Wales), Dr Absulsamie Hanano (Department of Molecular Biology and Biotechnology, AECS, Damascus, Syria) and Dr Bishnu Sarker (Meharry Medical College, Tennessee, USA).

Computer Science (8)

Funding Notes

No funding is available for this project


• Alastair M Kilpatrick, Farzana Rahman, et al. Characterizing domain-specific open educational resources by linking ISCB Communities of Special Interest to Wikipedia, Bioinformatics, Volume 38, Issue Supplement_1, July 2022, Pages i19–i27
• Vink, Guillaume, Nebel, Jean-Christophe and Wren, Stephen P (2021) In silico design of bioisosteric modifications of drugs for the treatment of diabetes. Future Medicinal Chemistry, 13(8), pp. 691-700. ISSN (print) 1756-8919
• Hanano, A., Perez-Matas, E., Shaban, M. et al. Characterization of lipid droplets from a Taxus media cell suspension and their potential involvement in trafficking and secretion of paclitaxel. Plant Cell Rep 41, 853–871 (2022).
• Ruvinga, Stenford, Hunter, Gordon J.A., Duran, Olga and Nebel, Jean-Christophe (2021) Use of LSTM neural networks to identify 'queenlessness' in honeybee hives from audio signals. In: 17th International Conference on Intelligent Environments; 21 - 24 Jun 2021, Dubai, United Arab Emirates.
• Guo, S., Vieweger, M., Zhang, K. et al. Ultra-thermostable RNA nanoparticles for solubilizing and high-yield loading of paclitaxel for breast cancer therapy. Nat Commun 11, 972 (2020).
• Bhattacharyya, J., Bellucci, J., Weitzhandler, I. et al. A paclitaxel-loaded recombinant polypeptide nanoparticle outperforms Abraxane in multiple murine cancer models. Nat Commun 6, 7939 (2015).