PhD2022-04 - Monitoring anti-microbial resistance reservoirs and the evolution of virulence through AI-supported next generation horizontal gene transfer annotation

Background

Microbial communities exchange genetic material not only through vertical inheritance but also through genetic recombination with other organisms. Those events, termed horizontal gene transfer (HGT), are for example major contributor to the acquisition of resistance to antibiotics (AMR). High-Throughput Sequencing (HTS) data is instrumental for the annotation of HGT in pathogens. Traditional HTS analysis methods build upon statistical techniques and algorithms to infer a set of the most likely predictions, provided the evidence from the sequences. However, the use of AI, in particular Deep Convolutional Neural Networks, can bypass parameter tuning and was shown to outperform state-of-the-art variant detection tools when applied to human genomes.

Aim

The aim of the project is to develop a variant prediction tool designed for the detection of horizontal gene transfer and monitor AMR reservoirs and the evolution of virulence.

AI Methods

You will develop a Deep Neural Network architecture that can accurately detect HGT directly from sequence data without the need for an expert to set up the system. High-throughput sequence data from public databases will be used to train the models. You will use Deep Neural Network as well as various supervised learning methods.

Apply via email here: [email protected]

nach oben