Papers
Research papers published in leading journals and accepted at renowned machine learning conferences.
Conferences
Papers from our Faculty members that have been accepted/presented at world-renowned AI and ML conferences.
International Conference on Learning Representations (ICLR)
Neural Information Processing Systems (NeurIPS)
International Conference on Machine Learning (ICML)
Biomedical Discovery
Scientists are currently hampered in the discovery of optimal drug targets by a failure to understand biological processes at a systems level.
Our aim in the biomedical arena is to leverage structural, metabolic and genetic metadata to develop multiscale generative models – at the level of molecules, cells, organs and whole organisms – to uncover novel targets for antibiotics and host-directed therapies.
We are also developing new, interpretable deep learning methods to better predict the functional impacts of the protein mutations seen in antimicrobial resistance and cancer.
Latest papers
Adversarial generation of gene expression data (Bioinformatics, February 2022)
Guest Editorial: Non-Euclidean Machine Learning (IEEE Transactions on Pattern Analysis and Machine Intelligence, February 2022)
An organoid CRISPRi screen revealed that SOX9 primes human fetal lung tip progenitors to receive WNT and RTK signals (bioRxiv, January 2022)
SCMTHP: A New Approach for Identifying and Characterizing of Tumor-Homing Peptides Using Estimated Propensity Scores of Amino Acids (Pharmaceutics, January 2022)
Voices on technology: The molecular biologists’ ever-expanding toy box (Molecular Cell, January 2022)
Early downregulation of hsa-miR-144-3p in serum from drug-naïve Parkinson’s disease patients (Nature: Scientific Reports, January 2022)
Ongoing research
Floto Group: The Floto Lab is looking at how immune cells interact with bacteria, how population-level whole genome sequencing can be used to reveal the biology of bacterial infection, and how therapeutic enhancement of cell-autonomous immunity may provide novel strategies to treat multi drug resistant pathogens.
Next-Generation Clinical Trials
Randomised controlled trials (RCTs) are currently considered the gold standard but such clinical trials are often slow, costly and lack flexibility.
We’re investigating how to optimise patient recruitment for enrolment in clinical trials and how to conduct more efficient, responsive trials. ML-enhanced trials can speed up learning and significantly reduce error.
Our work is informing the next generation of adaptive clinical trials, including determining sequential recruitment of patients, arriving at effective drug dosages and much more.
Precision Medicine
We are integrating omics data and a wide variety of other data sets, including electronic health records (EHRs), and applying novel machine learning techniques to better characterise individual patients, improve diagnosis, account for co-morbidities and reliably predict patient trajectories.
These technologies will enable the optimisation of targeted treatment interventions, identify looming health issues in large populations before they develop symptoms, and allow us to move towards the AI-enabled hospitals of tomorrow.
Latest papers
AI-Based Reconstruction for Fast MRI—A Systematic Review and Meta-Analysis (IEEE, February 2022)
Identifying healthy individuals with Alzheimer neuroimaging phenotypes in the UK Biobank (medRxiv, January 2022)
Conservative Policy Construction Using Variational Autoencoders for Logged Data With Missing Values (IEEE, January 2022)
A class-contrastive human-interpretable machine learning approach to predict mortality in severe mental illness (NPJ Schizophrenia, December 2021)
Disentangled Counterfactual Recurrent Networks for Treatment Effect Inference over Time (December 2021)
“Watch Me Grow- Electronic (WMG-E)” surveillance approach to identify and address child development, parental mental health, and psychosocial needs: study protocol (BMC Health Services Research, November 2021)
Omics Analytics
Our researchers are developing new machine-learning methods in omics analytics to discover the drivers of disease – at both the population and individual level.
Our techniques are allowing us to characterise clusters of cells or patients, enabling next-generation precision medicine that includes powerful predictive models of personalised disease trajectory, dynamic prognosis and likely responses to treatment.
Latest papers
The discovAIR project: a roadmap towards the Human Lung Cell Atlas (European Respiratory Journal, February 2022)
A human fetal lung cell atlas uncovers proximal-distal gradients of differentiation and key regulators of epithelial fates (bioRxiv, January 2022)
Expression Atlas update: gene and protein expression in multiple species (Nucleic Acids Research, January 2022)
Completing the cancer jigsaw puzzle with single-cell multiomics (Nature Cancer, December 2021)
MultiMAP: dimensionality reduction and integration of multimodal data (Genome Biology, December 2021)
COVID-19
The threat of coronavirus requires that we use every weapon in our scientific arsenal, from advances in fundamental science to capacity planning in intensive care departments and developing new approaches to clinical trials.
Our expert network brings pioneering AI and ML to bear on COVID-19.
Latest papers
Altered TMPRSS2 usage by SARS-CoV-2 Omicron impacts tropism and fusogenicity (Nature, February 2022)
Can we reliably automate clinical prognostic modelling? A retrospective cohort study for ICU triage prediction of in-hospital mortality of COVID-19 patients in the Netherlands (International Journal of Medical Informatics, January 2022)
Sounds of COVID-19: exploring realistic performance of audio-based digital testing (Nature Partner Journals: Digital Medicine, January 2022)
Obesity associated with attenuated tissue immune cell responses in COVID-19 (bioRxiv, January 2022)
Local and systemic responses to SARS-CoV-2 infection in children and adults (Nature, December 2021)
Ongoing research
van der Schaar Lab: The van der Schaar Lab is calling upon governments to adopt proven machine learning methods and use existing data to help healthcare infrastructure respond to the pandemic.
ML tools
Policy Impact Predictor (PIP): A machine learning tool developed by the van der Schaar Lab to guide government decision-making around measures to prevent the spread of COVID-19.