Research Associate in Machine-Learning Modelling of 2-D and 3-D Images (10406)

Research Associate in Machine-Learning Modelling of 2-D and 3-D Images (10406)

The Rosalind Franklin Institute (the Franklin) is a technology institute established by the UK Government as a unique centre committed to advancing tools that are needed to transform healthcare in the future.

The Institute brings together researchers in life and physical sciences, and engineering, to develop a spectrum of tools which we will use to image, interpret and intervene in biological systems.

These insights will speed up the discovery of new medicines, help find new diagnostics and contribute to a deeper understanding of human health and disease.

Our Science Strategy seeks to focus the Franklin’s research and unite our researchers around our Technology Innovation Challenges and Life Science Challenges.

For more information on the Franklin’s Challenges click here.

This is an exciting research opportunity to join the Virtu AI Cell project which aims to develop virtual mapping of intracellular contents of biological cells.

This leads to the ultimate goal of running in-silico cell simulations that use such mappings to model how the dynamics of intracellular contents may modulate in response to various types of cell disturbances; e. g., due to the injection of a drug molecule; or the intracellular growth of cancer promoting mutations, the checking of which one could model to predict new routes towards preventative cancer.

The Research Associate will focus on developing machine-learning (ML) approaches that automatically segment and classify cellular species from experimental 2-D microscopy and 3-D tomography data, to help build a virtual mapping of biological cells.

The role-holder will also explore new ways to represent data to more efficiently identify cellular species and enable better downstream ML modelling tasks.

The work may also involve designing, building and deploying new feature selection and optimisation strategies for the ML modelling of cellular data.

More generally, this role offers a great opportunity for an ML expert to become a valued member of a highly interdisciplinary, challenge-focused project team that aims to solve a global research challenge.

As a Research Associate at the Franklin, you will bring scientific knowledge and skills to deliver a specific research project and/or you will bring independent, creative science, or specific skills to a team delivering a project or program.

Through this work, you will build scientific independence, develop new science and leadership skills, and establish a growing reputation externally.

Key Responsibilities

• Analyse and interpret experimental datasets of 2-D microscopy images of cells via ML methods
• Analyse and interpret experimental datasets of 3-D tomograms of cellular data via ML methods
• Create data representations that help to develop efficient ML modelling of cellular data
• Perform feature selection and optimisation of sub-cellular data using ML modelling
• Construct, train, evaluate and deploy ML models for application to cell biology
• Curate large volumes of simulated data to assist with ML modelling for cell biology
• Undertake data cleaning, normalisation and splitting tasks as needed
• Perform or supervise data annotation where necessary
• Improve ML models via hyperparameter optimisation and/or fine-tuning parameters
• Integrate ML models into larger software systems, including live system architectures
• Develop automation tools that streamline ML-based predictions for cell biology
• Develop multi-fidelity approaches to ML modelling where needed
• Develop new ML algorithms for cell-biology applications where needed
• Plan data acquisitions with experimentalists who conduct them, analyse and interpret results and supervise delivery of outputs (e. g. research report, patent application) in a scientific/technology area of interest.
• Work within a project team, contributing to wider projects around key Challenges.
• Lead major contributions to outputs from research including papers, patents and both internal and external presentations.
• Support and develop others including day‑to‑day supervision of students or visitors in areas related to own research.
• Have supervised, staged progression to first stages of scientific independence with opportunities to further develop science and skills/experience.
• Enhance your research through collaboration with other researchers and make active contributions to exchanging ideas through your own network.
• Be able to understand, interpret, create and communicate appropriately within a research context.
• Develop search and discovery skills and techniques.
• Be supervised by a Scientist/Senior Scientist in delivery of research outputs, either in the context of a project or Challenge or as an early career development fellow.
• #J-18808-Ljbffr