CoreShell ML Protein Assembly PDRA

University of Strathclyde; Glasgow; On-site; GBP 42,000 - 52,000; Full time

Job summary

The University of Strathclyde is seeking a Research Associate to lead the machine-learning component of the CoreShell Fibre Foundry programme in the ARIA-funded project. The role involves building surrogate models that link protein sequences to experimental outcomes and guiding design-build-test cycles within an interdisciplinary team of computational chemists and engineers.

You will design features, develop Python workflows, and communicate results to collaborators.

Qualifications

• PhD in computational chemistry, chemical physics, molecular modelling, bioinformatics, machine learning, computational biology, materials informatics or a closely related discipline is required.
• Experience developing and evaluating machine-learning models is essential.
• Strong Python and scientific computing skills are required.
• Ability to work independently and as part of a collaborative team.

Responsibilities

• Develop machine-learning and active-learning models for predictive protein-assembly design.
• Define suitable molecular descriptors, input features and prediction targets.
• Analyse sequence, simulation and experimental datasets.
• Prioritise candidate protein sequences for simulation and experimental validation.
• Develop robust, documented Python workflows and share results with collaborators.
• Contribute to project meetings, milestone reports, and publications.

Skills

Machine-learning

Python

Data analysis

Active learning

Bayesian optimization

Education

PhD in a relevant field

Tools

Python workflows

Molecular descriptors

Job description

CoreShell ML Protein Assembly PDRA (845541)

FTE:1.0 FTE (full time)

Contract Type:Fixed term (18 Months)

Closing Date:25/09/2026

The Department of Pure and Applied Chemistry is seeking anexceptional and highly motivated Research Associate to join theARIA-funded CoreShell Fibre Foundry programme.

The programme aims to develop a new approach tomanufacturing hollow inorganic fibres using engineered proteinsas reusable molecular fabrication units. You will lead themachine-learning component of the computational workpackage, developing predictive and active-learning approachesto guide the design of protein sequences that assemble intocontrolled geometries.

You will build and evaluate surrogate models linking proteinsequence and molecular-simulation descriptors to experimentallyobserved assembly outcomes. These models will be used toprioritise candidate protein sequences for simulation andexperimental testing and will be refined through iterative design build-test cycles.

Working within an interdisciplinary team of computationalchemists, protein scientists and engineers, you will:

• develop machine-learning and active-learning models forpredictive protein-assembly design
• define suitable molecular descriptors, input features andprediction targets
• analyse sequence, simulation and experimental datasets
• prioritise candidate protein sequences for simulation andexperimental validation
• develop robust, documented and reproducible Pythonworkflows
• communicate model outputs clearly to computational andexperimental collaborators
• contribute to project meetings, milestone reports, publications,presentations and research-data management.

You will have a PhD in computational chemistry, chemicalphysics, molecular modelling, bioinformatics, machine learning,computational biology, materials informatics or a closely relateddiscipline. You will also have experience developing andevaluating machine-learning models, strong Python andscientific-computing skills, and the ability to work bothindependently and as part of a collaborative research team.

Experience of active learning, Bayesian optimisation, uncertaintyquantification, surrogate modelling, protein or peptide design,molecular descriptors, sequence-based modelling or moleculardynamics data would be advantageous.

The post is full time and fixed term from 1 October 2026 until 29February 2028. Any continuation beyond this initial funded periodwould be subject to successful passage of the programmemilestone gate, the release of further funding and separateUniversity approval.

Initial interviews have been scheduled for a date to be confirmed.

Informal enquiries may be directed to Professor Tell Tuttle,Programme Lead, at [e-mail]

CoreShell ML Protein Assembly PDRA (845541)