Recent Publications

Accelerate Science pursues research at the interface of AI and the sciences, generating new scientific insights and developing AI methods that can be deployed to advance scientific knowledge. This research is carried out in partnership with a community of scientists and AI specialists passionate about the use of AI to benefit science and society. Recent publications from the team are listed below.

Liquid Water Contains the Building Blocks of Diverse Ice Phases

Bartomeu Monserrat, Jan Gerit Brandenburg, Edgar A. Engel, Bingqing Cheng Nature Communications, 11 (5757):

This paper investigates the structural similarities between liquid water and a comprehensive set of 54 simulated ice phases. Its findings shed light on the phase behaviour of water.

Read more

Gaussian Process Latent Variable Flows for Massively Missing Data

Vidhi Lalchand, Aditya Ravuri, Neil D. Lawrence Third Symposium on Advances in Approximate Bayesian Inference,:

Gaussian process latent variable models (GPLVM) are used to perform nonlinear and probabilistic dimensionality reduction. This work develops a flexible model class, called Gaussian process latent variable flows (GPLVF), and evaluates its performance in massively missing data settings.

Read more

Quantum-Mechanical Exploration of the Phase Diagram of Water

Aleks Reinhardt, Bingqing Cheng Nature Communications, 12 (588):

This paper analyses the phase diagram of water at three hybrid density-functional-theory levels of approximation. It demonstrates that it is possible to predict the phase diagram of a polymorphic system from first principles and provides a thermodynamic way of testing the limits of quantum-mechanical calculations.

Read more

Optimal marker gene selection for cell type discrimination in single cell analyses

Bianca Dumitrascu, Soledad Villar, Dustin G. Mixon, Barbara Englehardt Nature Communications, 12 (1186):

This paper presents scGeneFit, an approach for marker selection in single cell RNA-seq. This method selects gene markers that jointly optimize cell label recovery using label-aware compressive classification methods.

Read more

Ranking the Information Content of Distance Measures

Aldo Glielmo, Claudio Zeni, Bingqing Cheng, Gabor Csanyi, Alessandro Laio:

Researchers can assess the similarity of different data points by creating measures of distance between them. This paper assesses the amount of information contained in different distance measures.

Read more

Solving Schrödinger Bridges via Maximum Likelihood

Francisco Vargas, Pierre Thodoroff, Austen Lamacraft, Neil D. Lawrence Entropy, 23 (9):1134

The Schrödinger bridge problem was proposed in the 1930s by Erwin Scrödinger. It involves two probability distributions at a start time an end time. They are related by a dynamic process. Deciding the form of that dynamic process given these distributions is a challenging problem. In this work we provide a maximum likelihood approach to solving it.

Read more

Deep learning for Bioimage Analysis in Developmental Biology

Adrien Hallou, Hannah G. Yevick, Bianca Dumitrascu, Virginie Uhlmann Development, 148 (18):

This review introduces key concepts in deep learning and its application to bio-image analysis, exploring how researchers can integrate these techniques into their work.

Read more

Phase Behaviours of Superionic Water at Planetary Conditions

Bingqing Cheng, Mandy Bethkenhagen, Chris J. Pickard, Sebastien Hamel Nature Physics, :

Understanding the thermodynamic and transport properties of water is crucial for planetary science. This paper uses machine learning to characterise the phase behaviours of water at extreme conditions.

Read more

Multimodal Graph Coarsening for Interpretable, MRI-Based Brain Graph Neural Network

Isaac Sebenius, Alexander Campbell, Sarah E. Morgan, Edward T. Bullmore, Pietro Liò IEEE 31st International Workshop on Machine Learning for Signal Processing (MLSP),:

This work applies graph neural networks to the study of MRI-based brain scans. It presents a model that classifies brain images from patients with schizophrenia and healthy control subjects, demonstrating that the use of multi-modal data can increase the accuracy of results.

Read more

Natural Language Processing markers in First Episode Psychosis and People at Clinical High-risk

Sarah E. Morgan, Kelly Diederen, Petra E. Vértes, Samantha H. Y. Ip, Bo Wang, Bethany Thompson, Arsime Demjaha, Andrea De Micheli, Dominic Oliver, Maria Liakata, Paolo Fusar-Poli, Tom J. Spencer, Philip McGuire Translational Psychiatry, 11 (630):

Disorganised speech can help predict later psychotic illness. This paper assesses the performance of twelve automated Natural Language Processing markers in differentiating transcribed speech excerpts from subjects at clinical high risk for psychosis, first episode psychosis patients and healthy control subjects.

Read more