Biology subjects

Griffie, J.

Publications and source records attributed to Griffie, J..

11 recordsLinked to original sources

bioRxiv2026

SMLMFlow: Improving Structural Resolution in Single Molecule Localization Microscopy with Flow Matching

While Single Molecule Localization Microscopy (SMLM) aims to generate precise coordinates of molecular targets in cells, the resulting point clouds are inherently blurred by additive noise sources across the experimental, imaging, and processing workflow. This blurring often limits SMLMs ability to accurately quantify complex assembled structures required to address biological issues, despite reported localization precision down to a couple of nanometers. Here, we present SMLMFlow, a machine learning framework for improving structural resolution in SMLM datasets that combines a graph neural network and a hierarchical transformer with flow matching. We show that SMLMFlow improves structural resolution and downstream quantification across different structures, including filaments and protein nano-clusters, and generalizes to new unseen photophysics models.

bioinformatics↗

bioRxiv2025

AI4CellFate: Interpretable Early Cell Fate Prediction with Generative AI

Live-cell imaging provides a unique insight into complex cellular processes including single cell fate, but remains limited by both low-throughput and the lack of generalisable analytics for the multidimensional datasets it produces. This work introduces AI4CellFate, an interpretable and data-driven machine learning framework for predicting cell fate from microscopy timelapses, applied here to cancer therapy. By integrating generative AI and contrastive learning, AI4CellFate enables early fate prediction as well as visualisation of biologically relevant features, with limited annotation.

bioinformatics↗

bioRxiv2025

Nanoscale protein clustering modulates the input-output response of cellular signalling pathways

A key function of the plasma membrane is regulating signal transduction. Proteins involved in the transduction process play a signal-processing role; mapping an input (e.g. number of engaged receptors) to an output (e.g. level of downstream phosphorylation). In many cases, a digital mapping is desirable, i.e. that a cell activates or responds fully once a set input threshold is surpassed. It is believed that the nanoscale organisation of proteins, such as their clustering, modulates this behaviour by altering the frequency of protein-protein interactions. Here, we develop a generalised simulator for dynamic molecular clustering built around agent-based modelling. We show that the clustering properties (e.g. size of clusters, percentage of monomers) tunes the cellular response. This work paves the way for designing potential therapeutic interventions that alter the nanoscale patterning of molecules on the cell surface in order to alter cell behaviour. AUTHOR SUMMARYSignal transduction is the process by which cells convert external signals into internal responses. This is essential for such functions as hormone signalling and immunity. Many membrane proteins form nanoscale clusters on the cell surface, and state-of-the-art imaging methods allow us to see these clusters in detail. We developed an agent-based simulator to explore how clustering of key signaling proteins (such as kinases and phosphatases) can influence cell signalling outputs. Our results suggest that protein clustering can "digitize" signals, pushing cells to respond in an "all-or-nothing" manner, rather than producing a smooth range of responses. This research contributes to a broader understanding of how dynamic reorganization of transmembrane proteins contributes to information processing. This is a central question in systems biology, with relevance to the fields of immunology, neuroscience, and cancer. By highlighting how the geometric properties of protein distributions can alter signalling responses, we may identify novel methods to manipulate cell behavior.

systems biology↗

bioRxiv2024

An unsupervised deep learning framework encodes super-resolved image features to decode bacterial cell cycle

Super-resolution microscopy can resolve cellular features at the nanoscale. However, increased spatial resolution comes with increased phototoxicity, and reduced temporal resolution. As a result, studies that require the highest spatial resolutions often rely on static or fixed images, lacking dynamic information. This is particularly true of bacteria, whose lateral dimensions approach the scale of the diffraction limit. In this work, we present Enso, a method based on unsupervised machine learning to recover bacterial cell cycle and cell type information from static single molecule localization microscopy (SMLM) images, whilst retaining nanoscale spatial resolution. Enso uses single-cell images as input, and orders cells according to their spatial pattern progression, ultimately linked to the cell cycle. Our method requires no a priori knowledge or categories, and is validated on both simulated and user-annotated experimental data.

bioinformatics↗

bioRxiv2021

Event-driven acquisition for content-enriched microscopy

A common goal of fluorescence microscopy is to collect data on specific biological events. Yet, the event-specific content that can be collected from a sample is limited, especially for rare or stochastic processes. This is due in part to photobleaching and phototoxicity, which constrain imaging speed and duration. We developed an event-driven acquisition (EDA) framework, in which neural network-based recognition of specific biological events triggers real-time control in an instant structured illumination microscope (iSIM). Our setup adapts acquisitions on-the-fly by switching between a slow imaging rate while detecting the onset of events, and a fast imaging rate during their progression. Thus, we capture mitochondrial and bacterial divisions at imaging rates that match their dynamic timescales, while extending overall imaging durations. Because EDA allows the microscope to respond specifically to complex biologi-cal events, it acquires data enriched in relevant content.

cell biology↗

bioRxiv2021

A framework for evaluating the performance of SMLM cluster analysis algorithms

Single molecule localisation microscopy (SMLM) generates data in the form of Cartesian coordinates of localised fluorophores. Cluster analysis is an attractive route for extracting biologically meaningful information from such data and has been widely applied. Despite the range of developed cluster analysis algorithms, there exists no consensus framework for the evaluation of their performance. Here, we use a systematic approach based on two metrics, the Adjusted Rand Index (ARI) and Intersection over Union (IoU), to score the success of clustering algorithms in diverse simulated clustering scenarios mimicking experimental data. We demonstrate the framework using three analysis algorithms: DBSCAN, ToMATo and KDE, show how to deduce optimal analysis parameters and how they are affected by fluorophore multiple blinking. We propose that these standard conditions and metrics become the basis for future analysis algorithm development and evaluation.

biophysics↗

bioRxiv2021

Correction of multiple-blinking artefacts in photoactivated localisation microscopy

Photoactivated localisation microscopy (PALM) produces an array of localisation coordinates by means of photoactivatable fluorescent proteins. However, observations are subject to fluorophore multiple-blinking and each protein is included in the dataset an unknown number of times at different positions, due to localisation error. This causes artificial clustering to be observed in the data. We present a workflow using calibration-free estimation of blinking dynamics and model-based clustering, to produce a corrected set of localisation coordinates now representing the true underlying fluorophore locations with enhanced localisation precision. These can be reliably tested for spatial randomness or analysed by other clustering approaches, and previously inestimable descriptors such as the absolute number of fluorophores per cluster are now quantifiable, which we validate with simulated data. Using experimental data, we confirm that the adaptor protein, LAT, is clustered at the T cell immunological synapse, with its nanoscale clustering properties depending on location and intracellular phosphorylatable tyrosine residues.

biophysics↗

bioRxiv2020

Virtual-SMLM, a virtual environment for real-time interactive SMLM acquisition

Although single molecule localisation microscopy enables for the visualisation of cells nanoscale organisation, its dissemination remains limited mainly due to the complexity of the associated imaging acquisition, impacting on outputs reliability and reproducibility. We propose here the first all-in-one fully virtual environment for SMLM acquisition: Virtual-SMLM, including on-the-fly interactivity and real time display. It relies on a novel realistic approach to simulate fluorophores photo-physics based on independent pseudo-continuous emission traces. It also facilitates for user-specific experimental and optical environment design. As such, it constitutes a unique tool for the training of both users and machine learning approaches to automated SMLM, as well as for experimental validation, whilst providing realistic data sets for the development of image reconstruction algorithms and data analysis software.

biophysics↗

bioRxiv2019

An agent-based model of molecular aggregation at the cell membrane

Molecular clustering at the plasma membrane has long been identified as a key process and is associated with regulating signalling pathways across cell types. Recent advances in microscopy, in particular the rise of super-resolution, have allowed the experimental observation of nanoscale molecular clusters in the plasma membrane. However, modelling approaches capable of recapitulating these observations are in their infancy, partly because of the extremely complex array of biophysical factors which influence molecular distributions and dynamics in the plasma membrane. We propose here a highly abstracted approach: an agent-based model dedicated to the study of molecular aggregation at the plasma membrane. We show that when molecules are modelled as though they can act (diffuse) in a manner which is influenced by their molecular neighbourhood, many of the distributions observed in cells can be recapitulated, even though such sensing and response is not possible for real membrane molecules. As such, agent-based offers a unique platform which may lead to a new understanding of how molecular clustering in extremely complex molecular environments can be abstracted, simulated and interpreted using simple rules.\n\nAuthor summaryMolecular aggregation in cell membranes is a key component of cellular machinery, involved across cell types in inter-cellular communication and signalling pathway initiation. As such, understanding the underlying mechanisms and molecule cluster characteristics at a more theoretical level is a pre-requisite. Complete descriptive molecular models have proven impossible to realise due to the overall complexity of the processes involved, highlighting the need for novel approaches. While conceptual models have been shown to be powerful tools and are routinely used in other fields with high level of complexity such as social sciences or economics, they are overall lacking from the literature when it comes to cell studies. We suggest in this work that the same principle applies to cell biology and in particular, the study of molecular clustering. We propose here a general model, independent of cell types or signalling pathways: an agent-based model dedicated to molecular clustering in the plasma membrane. We show we are able to recapitulate molecular aggregation similar to observations in cells while new properties are highlighted by our model, for instance, clustering is a digitised process.

bioinformatics↗

bioRxiv2019

Differential nanoscale organisation of LFA-1 modulates T cell migration

Effector T-cells rely on integrins to drive adhesion and migration to facilitate their immune function. Heterodimeric transmembrane integrin LFA-1 (L{beta}2) regulates adhesion and migration through linkage of the extracellular matrix with the intracellular actin treadmill machinery. We quantitated the velocity and direction of F-actin flow in migrating T-cells alongside single molecule localisation of transmembrane and intracellular LFA-1. Our results show that retrograde actin flow positively correlated and immobile actin negatively correlated with T-cell velocity. Plasma membrane localised LFA-1 forms unique nano-clustering patterns in the leading edge, compared to the mid-focal zone, in migrating T-cells. Deleting the cytosolic phosphatase PTPN22, a negative regulator of integrin signaling, increased T-cell velocity, and leading-edge cluster co-localisation of pY397 FAK, pY416 Src family kinases and LFA-1. These data suggest that differential nanoclustering patterns of LFA-1 in migrating T-cells can instruct intracellular signalling linked with the actin treadmill. Our data presents a paradigm where T cells modulate the nanoscale organisation of adhesion and signalling molecules to fine tune their migration speed. This has implications for the regulation of immune and inflammatory responses.

immunology↗

bioRxiv2018

Machine-learning for cluster analysis of localization microscopy data.

Quantifying the clustering of points within single-molecule localization microscopy data is useful to understanding the spatial relationships of the molecules in the underlying sample. The conversion of point pattern data into a meaningful description of clustering is difficult, especially for biologically derived data, as the definitions of clustering are often subjective or simplistic. Many existing computational approaches are also limited in their ability to process large-scale data-sets or to deal effectively with inhomogeneities in clustering. Here we have developed a supervised machine-learning approach to cluster analysis which is fast and accurate. Trained on a variety of simulated clustered data, the network can then classify all points from a typical localization microscopy data-set (several million points from the entire field of view) as being either clustered or not-clustered, with the potential to include additional classifiers to describe different types of clusters. Clustered points can then be further refined into like-clusters for the measurement of cluster area, shape, and point-density. We demonstrate the performance on simulated data and experimental data of the kinase Csk and the adaptor PAG in both naive and pre-stimulated primary human T cell synapses.

cell biology↗