Deep learning for optimisation of cell facto

Lead Research Organisation: University of Edinburgh
Department Name: Sch of Biological Sciences

Abstract

Biotechnology is a $1tn industry poised to help against global challenges in the health, food and energy sectors. One of the most promising biotechnologies are "cell factories", where genetically engineered microbes are employed for converting cheap feedstocks into high-value chemicals. At present, however, cell factory development relies on many trial-and-error rounds of design and testing, which slows down R&D and stifles innovation.

This project aims to develop cutting-edge artificial intelligence methods for optimisation of protein production in living cells. The central challenge lies in the difficulty of predicting expression from DNA sequence, known as the 'genotype-phenotype mapping' problem that has puzzled scientists for over half a century.

In the project you will develop advanced deep learning pipelines to forecast production and radically improve strain performance. The algorithms will predict protein production from DNA sequence alone, providing a unique, data-driven, approach for accelerating strain design and optimisation. We will build a suite of algorithms for finding DNA sequences that maximise protein production in industrially-relevant hosts. To this end you will use large genotype-phenotype screens from the literature in combination with machine learning and optimization methods, to build accurate predictors of protein expression and efficient algorithms to discover super-producing strains.

The student will join the labs of Diego Oyarzún and Karl Burgess, who combine expertise in synthetic biology, machine learning, optimisation, metabolomics, and strain characterisation. The ideal candidate should have excellent mathematical and computational skills, and must hold a First Class or an Upper Second Class degree (or equivalent overseas qualification) in a discipline relevant to the project such as molecular biology, bioinformatics, mathematics, or engineering. The successful applicant will benefit from a diverse and multidisciplinary team at the interface of biology and computation, with multiple opportunities for training, networking and career growth.

Publications

10 25 50

Studentship Projects

Project Reference Relationship Related To Start End Student Name
BB/T00875X/1 30/09/2020 29/09/2028
2744281 Studentship BB/T00875X/1 30/09/2022 29/09/2026