The Turing Podcast

This week's episode takes us back the pre-covid19 era, when premier league football was still a thing! Dr Nick Barlow of The Alan Turing Institute's Research Engineering Group chats to the podcast hosts about his side project "AIrsenal", a machine learning manager for Fantasy Premier League. You can find out more about the project and even contribute to the code via these links:

- https://www.turing.ac.uk/news/airsenal-difficult-second-season

- https://github.com/alan-turing-institute/AIrsenal/blob/master/CONTRIBUTING.md

Viruses like SARS-CoV-2, which causes Covid-19, aren't the only problem we face globally when it comes to infectious disease. Bacterial pathogens are constantly evolving, presenting a challenge for the worlds healthcare systems as more of the antibiotics we rely on to kill bacteria fail on those that have evolved antimicrobial resistance. In this episode, we chat with Victoria Carr, a PhD student in Bioinformatics at the Centre for Host-Microbiome Interactions Kings College London and The Alan Turing Institute, who describes her research comparing antimicrobial resistance in mouth and gut microbes and developing software to find "mobile genetic elements" (DNA sequences that can change position within a genome) and their association with antibiotic resistance genes. Be sure to also check out Vicky's podcast about women in tech at  https://www.researcherscode.com/podcast

 

Image recognition and classification is a hot topic in AI research, and these tools are increasingly being utilised by biologists with the aim to classify and distinguish diseases. In this episode, Bea Costa Gomes, a PhD candidate and Turing enrichment student at The University of Manchester talks about her research into developing software that spots shape differences in the images of diseased brain cells, as well as her passion for Drosophila flies!

Remember back at school when you were taught that correlation doesn’t mean causation, that increased ice cream sales are correlated with sunnier weather but don’t cause the clouds to part? Peter Tennant, a fellow of the Alan Turing Institute based at Leeds Institute for Data Analytics explains why it’s important for scientists to become more confident in talking about causation, how "causal inference" methods are transforming the field of epidemiology and why AI isn’t typically best placed to make sensible assumptions about complex data. This episode was recorded before the Covid-19 lockdown began in the UK, but the topics discussed couldn’t be more relevant!

Large datasets and modern machine learning techniques are fast changing the field of Astrophysics and our understanding of the universe. In this episode, we chat with PhD candidates Tarek Allam and Gordon Yip, whose respective research at University College London’s Department of Physics and Astronomy focuses on the classification of supernova light curves to help constrain theories of Dark Energy and the atmospheric composition of remote exo-planets.

How can groups of people be empowered to share sensitive personal data such as medical records, with transparency about how they will be used? This is a question of particular relevance during the Covid-19 pandemic, when scientific research into the disease can greatly benefit from data collection at scale. In this episode, we're talking to Sylvie Delacroix, Turing Fellow and Professor in Law and Ethics at Birmingham Law School (University of Birmingham) about the concept of data trusts and how they enable the sharing and safeguarding of data. https://datatrusts.uk/

How can data science support us in tackling the Covid-19 pandemic? Our Researcher at Large, Jon Crowcroft, speaks about how smartphones can be mobilised to track the Covid-19 pandemic. He also discusses how data science and AI research is playing a crucial role in finding solutions during the crisis.

 

Recorded a few months ago, in our first episode we speak to Dr Radka Jersakova, Research Data Scientist at The Institute about her project that applies AI to air traffic control simulators.  Join us as we discuss how advances in AI may go some way paving the way towards putting more planes (safely) in the sky.