ANC Workshop - 22/04/2025

Speaker: Domas Linkevicius, ANC PhD student 

Title: One model to rule them all: unification of voltage-gated potassium channel models via deep non-linear mixed effects modellingAbstract: Ion channels are essential for signal processing and propagation in neural cells. Voltage-gated ion channels permeable to potassium (Kv) form one of the most prominent channel families. Techniques used to model the voltage-dependent gating of Kv channels date back to Hodgkin and Huxley (Hodgkin, A. L., & Huxley, A. F. (1952)  JPhysiol (Lond), 117,500–544). Different Kv types can display radically different kinetic properties, requiring different mathematical models. However, the construction of Hodgkin-Huxley-like (HH-like) models is generally complex and time consuming due to the number of parameters, their tuning and having to choose functional forms to model gating. 

In addition to the between-Kv type heterogeneity, there can be significant within-Kv type kinetic heterogeneity between different cells with genetically identical channels, recently systematically measured by Ranjan et al. (Ranjan, R., et al. (2019) Frontiers in Cellular Neuroscience, 13. https://doi.org/10.3389/fncel.2019.00358). Since HH-like models do not account for such variability, extensions to it are necessary. We use scientific machine learning (SciML), the integration of machine learning methodologies with existing scientific models, and non-linear mixed effects (NLME) modelling to bypass the limitations of HH-like modelling. NLME is a modelling methodology that takes into account both within- and between-subject variability. These tools allowed us to complement the HH-like modelling and construct a unified SciML HH-like model that fits the recordings from 20 different Kv types. The unified SciML HH-like model produced closer fits to the data compared to a set of seven previous HH-like models and was able to represent the highly heterogeneous data from Ranjan et al. well. Our model may be the first step in producing a general foundation model for ion channels that would be capable of modelling the gating of any ion channel type.

Speaker: Owen Pauptit, UG4 CS and Maths project student

Title: Ethnic bias in England and Wales stop and search: can it be explained by relative amount of crime?

Abstract: Stop and search practice has been controversial for being prone to ethnic bias in England and Wales for a long time. There is a wealth

of literature on the topic demonstrating that ethnic minorities are stopped more than White people proportionally compared to relative

population sizes, but much of the existing research fails to account for essential confounding variables such as type of crime, location

and temporal trends when performed on a national scale. A major criticism of these studies is that discrepancies in stop and search

are simply a reflection of discrepancies in the amount of crime committed by each ethnic group.

We analysed data taken from the UK government for all 43 police forces across England and Wales from financial years 2015 to 2023,

including 4,037,331 stops and searches and 6,385,855 arrests in total. Inspired by the work of Gelman et al. (2007) on New York City

data, we use the number of arrests of each ethnic group as a proxy for amount of crime and fit a Bayesian hierarchical model to

estimate the rate of stops and searches compared to arrests for each ethnic group, whilst controlling for variation associated with

crime type, police force and financial year. We then extend this analysis by examining how this rate changes over time and producing

excursion sets of police forces: the largest set of police forces for which there is 95% joint probability that every police force in the set

stops and searches a particular ethnic minority more than White people compared to arrests. We find that Lancashire, North Yorkshire

and the West Midlands police forces all stop and search Asian people more than White people compared to arrests, regardless of year or crime type, with at least 95% joint probability.