Events

CDT hosts a variety of events throughout the year.

Sign up to our mailing list to get notified of our upcoming events.

Join CDT mailing list

Past events

Biomedical AI CDT seminar, 21 April 2021

Speaker: Chris Williams, School of Informatics

The practical work of deploying a machine learning system is dominated by issues outside of training a model: data preparation, data cleaning, understanding the data set, debugging models, and so on. The goal of the Artificial Intelligence for Data Analytics project at the Alan Turing Institute is to help to automate the whole data analytics process by drawing on advances in AI and machine learning. We will describe tools to address such tasks, including identifying syntactic and semantic data types, data integration, and identifying and repairing missing and anomalous data. Joint work with the AIDA team: Taha Ceritli, James Geddes, Ernesto Jimenez-Ruiz, Ian Horrocks, Alfredo Nazabal, Tomas Petricek, Charles Sutton, Gerrit Van Den Burg.

Video: Artificial Intelligence for Data Analytics
Speaker: Chris Williams, School of Informatics Abstract: The practical work of deploying a machine learning system is dominated by issues outside of training a model: data preparation, data cleaning, understanding the data set, debugging models, and so on. The goal of the Artificial Intelligence for Data Analytics project at the Alan Turing Institute is to help to automate the whole data analytics process by drawing on advances in AI and machine learning. We will describe tools to address such tasks, including identifying syntactic and semantic data types, data integration, and identifying and repairing missing and anomalous data. Joint work with the AIDA team: Taha Ceritli, James Geddes, Ernesto Jimenez-Ruiz, Ian Horrocks, Alfredo Nazabal, Tomas Petricek, Charles Sutton, Gerrit Van Den Burg.

Biomedical AI CDT seminar, 14 April 2021

Speaker: Jennifer Quint, National Heart and Lung Institute, Imperial College London & BREATHE Hub.

This talk will consider what clinicians think about data and what they should think about data. I’ll talk about how and where data are collected, used and how we can make it even better. Whilst clinicians use and report data in day-to-day clinical practice, entry tools and reporting are often inefficient, and whilst clinicians are skilled at observation and medical decision making, they often struggle with accurately recording and measuring activities. This in turn leads to data variation and issues in data quality.

Video: Data: A Clinical Perspective
Speaker: Jennifer Quint, National Heart and Lung Institute, Imperial College London & BREATHE Hub This talk will consider what clinicians think about data and what they should think about data. I’ll talk about how and where data are collected, used and how we can make it even better. Whilst clinicians use and report data in day-to-day clinical practice, entry tools and reporting are often inefficient, and whilst clinicians are skilled at observation and medical decision making, they often struggle with accurately recording and measuring activities. This in turn leads to data variation and issues in data quality.

 

Biomedical AI CDT seminar, 3 February 2021

Bruce Guthrie, Professor of General Practice at Usher Institute and Director of Advanced Care Research Centre

Sohan Seth, Senior Data Scientist, School of Informatics

Jacques Fleuriot, Director of AIAI institute, School of Informatics

Populations are rapidly ageing internationally, with the number of older people in the UK predicted to double in the next decade. Even before COVID-19, health and social care systems were already struggling to provide personalised care, and to effectively support people to live as independently as possible for as long as possible. This demographic transition is a major societal challenge, and new technologies and data-driven innovation are a key element of meeting the challenges posed.

The Advanced Care Research Centre is an interdisciplinary collaboration to research care in later life, and includes academics from seven schools in the University (Medicine, Informatics, Engineering, Health in Social Science, Literature Languages and Culture, Social and Political Science, and Art and Architecture).

The presentation will introduce the wider context of population ageing and care, and then discuss elements of the programme including (1) Analysis and prediction of later life joint trajectories of health, function, independence and death; and (2) New technologies of care such as passive and active sensing and other supportive tools to support independence and safety.

Video: Advanced Care Research Centre: interdisciplinary research in later life care
Populations are rapidly ageing internationally, with the number of older people in the UK predicted to double in the next decade. Even before COVID-19, health and social care systems were already struggling to provide personalised care, and to effectively support people to live as independently as possible for as long as possible. This demographic transition is a major societal challenge, and new technologies and data-driven innovation are a key element of meeting the challenges posed. The Advanced Care Research Centre is an interdisciplinary collaboration to research care in later life, and includes academics from seven schools in the University (Medicine, Informatics, Engineering, Health in Social Science, Literature Languages and Culture, Social and Political Science, and Art and Architecture). The presentation will introduce the wider context of population ageing and care, and then discuss elements of the programme including (1) Analysis and prediction of later life joint trajectories of health, function, independence and death; and (2) New technologies of care such as passive and active sensing and other supportive tools to support independence and safety.

 

 

Biomedical AI CDT seminar, 25 November 2020

Speaker: Dr Shilpa Garg, Harvard Medical School & Dana-Farber Cancer Institute

Reconstructing the complete phased sequences of every chromosome copy in human and non-human species are important for medical, population and comparative genetics. The unprecedented advancements in sequencing technologies have opened up new avenues to reconstruct these phased sequences that would enable a deeper understanding of molecular, cellular and developmental processes underlying complex diseases. Despite these interesting sequencing innovations, the highly polymorphic and gene-dense regions human leukocyte antigen (HLA) are not yet fully phased in the reference genome. The reference genome still contains gaps in multi-megabase repetitive regions, and thus annotating novel expression and methylation results are incomplete and inaccurate, that affect the interpretation of molecular genetics and epigenetics of diseases. There is a pressing need for a streamlined, production-level, easy-to-use computational algorithmic approaches that can reconstruct high-quality chromosome-scale phased sequences, and that can be applied to hundreds of human genomes.

In this talk, first, I will present a combinational optimization formulation and solution to the haplotype reconstruction problem that leverages new long-range Strand-specific technology and long reads to generate chromosome-scale phasing. Second, I present an efficient graph-based algorithm to perform accurate haplotype-resolved assembly of human individuals. The advantage of graphs is that they enable a unique compact representation of massive datasets for their integration on the common genome sequence space. This method takes advantage of new long accurate data type (PacBio HiFi) and long-range Hi-C data. We for the first time can generate accurate chromosome-scale phased assemblies with base-level-accuracy of Q50 and continuity of 25Mb within 24 hours per sample, therefore, setting up a milestone in the genomic community. Third, I will present the generalized computational approach that has the advantage to work on any type of sequencing data types for different number haplotypes and repeat variation. Finally, I will present the importance of haplotype-resolved assemblies to various medical applications.  In summary, my works develop scalable computational approaches that efficiently and robustly combine data from a variety of sequencing technologies to produce high-quality diploid assemblies. These computational methods have the potential to enable high-quality precision medicine and facilitate new and unbiased studies of human (and non-human) haplotype variation in various populations which are currently goals of the Human Genome Reference Project.

Video: Advanced computational approaches for understanding allele-specific biology of complex diseases
Reconstructing the complete phased sequences of every chromosome copy in human and non-human species are important for medical, population and comparative genetics. The unprecedented advancements in sequencing technologies have opened up new avenues to reconstruct these phased sequences that would enable a deeper understanding of molecular, cellular and developmental processes underlying complex diseases. Despite these interesting sequencing innovations, the highly polymorphic and gene-dense regions human leukocyte antigen (HLA) are not yet fully phased in the reference genome. The reference genome still contains gaps in multi-megabase repetitive regions, and thus annotating novel expression and methylation results are incomplete and inaccurate, that affect the interpretation of molecular genetics and epigenetics of diseases. There is a pressing need for a streamlined, production-level, easy-to-use computational algorithmic approaches that can reconstruct high-quality chromosome-scale phased sequences, and that can be applied to hundreds of human genomes. In this talk, first, I will present a combinational optimization formulation and solution to the haplotype reconstruction problem that leverages new long-range Strand-specific technology and long reads to generate chromosome-scale phasing. Second, I present an efficient graph-based algorithm to perform accurate haplotype-resolved assembly of human individuals. The advantage of graphs is that they enable a unique compact representation of massive datasets for their integration on the common genome sequence space. This method takes advantage of new long accurate data type (PacBio HiFi) and long-range Hi-C data. We for the first time can generate accurate chromosome-scale phased assemblies with base-level-accuracy of Q50 and continuity of 25Mb within 24 hours per sample, therefore, setting up a milestone in the genomic community. Third, I will present the generalized computational approach that has the advantage to work on any type of sequencing data types for different number haplotypes and repeat variation. Finally, I will present the importance of haplotype-resolved assemblies to various medical applications. In summary, my works develop scalable computational approaches that efficiently and robustly combine data from a variety of sequencing technologies to produce high-quality diploid assemblies. These computational methods have the potential to enable high-quality precision medicine and facilitate new and unbiased studies of human (and non-human) haplotype variation in various populations which are currently goals of the Human Genome Reference Project.