Search

Your search keyword '"McLean, Callum"' showing total 17 results
Start Over Author "McLean, Callum"
17 results on '"McLean, Callum"'

1. $\texttt{MiniMol}$: A Parameter-Efficient Foundation Model for Molecular Learning

Author

Kläser, Kerstin, Banaszewski, Błażej, Maddrell-Mander, Samuel, McLean, Callum, Müller, Luis, Parviz, Ali, Huang, Shenyang, and Fitzgibbon, Andrew

Subjects
Computer Science - Machine Learning, Computer Science - Artificial Intelligence
Abstract

In biological tasks, data is rarely plentiful as it is generated from hard-to-gather measurements. Therefore, pre-training foundation models on large quantities of available data and then transfer to low-data downstream tasks is a promising direction. However, how to design effective foundation models for molecular learning remains an open question, with existing approaches typically focusing on models with large parameter capacities. In this work, we propose $\texttt{MiniMol}$, a foundational model for molecular learning with 10 million parameters. $\texttt{MiniMol}$ is pre-trained on a mix of roughly 3300 sparsely defined graph- and node-level tasks of both quantum and biological nature. The pre-training dataset includes approximately 6 million molecules and 500 million labels. To demonstrate the generalizability of $\texttt{MiniMol}$ across tasks, we evaluate it on downstream tasks from the Therapeutic Data Commons (TDC) ADMET group showing significant improvements over the prior state-of-the-art foundation model across 17 tasks. $\texttt{MiniMol}$ will be a public and open-sourced model for future research.

Published
2024

2. Towards Foundational Models for Molecular Learning on Large-Scale Multi-Task Datasets

Author

Beaini, Dominique, Huang, Shenyang, Cunha, Joao Alex, Li, Zhiyi, Moisescu-Pareja, Gabriela, Dymov, Oleksandr, Maddrell-Mander, Samuel, McLean, Callum, Wenkel, Frederik, Müller, Luis, Mohamud, Jama Hussein, Parviz, Ali, Craig, Michael, Koziarski, Michał, Lu, Jiarui, Zhu, Zhaocheng, Gabellini, Cristian, Klaser, Kerstin, Dean, Josef, Wognum, Cas, Sypetkowski, Maciej, Rabusseau, Guillaume, Rabbany, Reihaneh, Tang, Jian, Morris, Christopher, Koutis, Ioannis, Ravanelli, Mirco, Wolf, Guy, Tossou, Prudencio, Mary, Hadrien, Bois, Therence, Fitzgibbon, Andrew, Banaszewski, Błażej, Martin, Chad, and Masters, Dominic

Subjects
Computer Science - Machine Learning
Abstract

Recently, pre-trained foundation models have enabled significant advancements in multiple fields. In molecular machine learning, however, where datasets are often hand-curated, and hence typically small, the lack of datasets with labeled features, and codebases to manage those datasets, has hindered the development of foundation models. In this work, we present seven novel datasets categorized by size into three distinct categories: ToyMix, LargeMix and UltraLarge. These datasets push the boundaries in both the scale and the diversity of supervised labels for molecular learning. They cover nearly 100 million molecules and over 3000 sparsely defined tasks, totaling more than 13 billion individual labels of both quantum and biological nature. In comparison, our datasets contain 300 times more data points than the widely used OGB-LSC PCQM4Mv2 dataset, and 13 times more than the quantum-only QM1B dataset. In addition, to support the development of foundational models based on our proposed datasets, we present the Graphium graph machine learning library which simplifies the process of building and training molecular machine learning models for multi-task and multi-level molecular datasets. Finally, we present a range of baseline results as a starting point of multi-task and multi-level training on these datasets. Empirically, we observe that performance on low-resource biological datasets show improvement by also training on large amounts of quantum data. This indicates that there may be potential in multi-task and multi-level training of a foundation model and fine-tuning it to resource-constrained downstream tasks.

Published
2023

4. Biomechanics and functional morphology of amblypygid predation

Author

McLean, Callum

Subjects
595.4
Abstract

Amblypygids, colloquially known as whip spiders, are a charismatic order of arachnids that a characterised a unique pair of spined pedipalp appendages. Amblypygid pedipalps are hypothesised to primarily function as a prey capture device. However, the pedipalp is also used in several other functions including territorial contest and courtship, opening the possibility that the appendage could also be under the influence of sexual selection. There exists a vast degree of morphological diversity within the pedipalp, with relative length spanning nearly an order of magnitude across the group and spination varying markedly both within- and between species. The amblypygid pedipalp is therefore subject to multiple selective pressures, and both its external morphology and kinematics likely reflect this. Thus, the amblypygid pedipalp provides an ideal structure through which to study the evolution of morphological traits subject to multiple selective pressures, and the potential evolutionary trade-offs that may arise. Despite this, amblypygid pedipalp morphology and kinematics remains poorly quantified and little comparative work has been carried across the group. Here, I aim to quantify intra- and interspecific trends in amblypygid pedipalp shape and prey capture kinematics for the first time, using modern morphometric techniques and high-speed videography. In this work I present a comprehensive review of sexual dimorphism in arachnids and identify the common drivers behind this phenomenon. Building on this, I quantify sexual dimorphism in pedipalp size and shape in a single species of amblypygid using a novel geometric morphometric approach. This is followed by an broad analysis of intraspecific and interspecific trends in shape complexity of pedipalps using Elliptical Fourier Analysis. Finally, I draw a link between form and function by quantifying and comparing prey capture kinematics in a morphologically diverse set of amblypygids, using high-speed videography and motion analysis. This work provides new insights into amblypygid 8 pedipalp diversity and posits the possibility of an evolutionary trade-off between increased pedipalp length, for use in display in courtship and territorial contest, and strike performance during prey capture.

Published
2020

5. The kinematics of amblypygid (Arachnida) pedipalps during predation: extreme elongation in raptorial appendages does not result in a proportionate increase in reach and closing speed

Author

McLean, Callum J, Brassey, Charlotte A, Seiter, Michael, Garwood, Russell J, Gardiner, James D, McLean, Callum J, Brassey, Charlotte A, Seiter, Michael, Garwood, Russell J, and Gardiner, James D

Abstract

The link between form and function is key to understanding the evolution of unique and/or extreme morphologies. Amblypygids, or whip spiders, are arachnids that often have highly elongated spined pedipalps. These limbs are used to strike at, and secure, prey before processing by the chelicerae. Amblypygi pedipalps are multifunctional, however, being used in courtship and contest, and vary greatly in form between species. Increased pedipalp length may improve performance during prey capture, but length could also be influenced by factors including territorial contest and sexual selection. Here for the first time, we use high-speed videography and manual tracking to investigate kinematic differences in prey capture between amblypygid species. Across six morphologically diverse species, spanning four genera and two families, we create a total dataset of 86 trials (9-20 per species). Prey capture kinematics varied considerably between species, with differences being expressed in pedipalp joint angle ranges. In particular, maximum reach ratio did not remain constant with total pedipalp length, as geometric scaling would predict, but decreased with longer pedipalps. This suggests that taxa with the most elongated pedipalps do not deploy their potential length advantage to proportionally increase reach. Therefore, a simple mechanical explanation of increased reach does not sufficiently explain pedipalp elongation. We propose other factors to help explain this phenomenon, such as social interactions or sexual selection, which would produce an evolutionary trade-off in pedipalp length between prey capture performance and other behavioural and/or anatomical pressures.

Published
2024

7. Assessing the patterns and drivers of shape complexity in the amblypygid pedipalp

Author

McLean, Callum, Garwood, Russell, Brassey, Charlotte, McLean, Callum, Garwood, Russell, and Brassey, Charlotte

Abstract

Amblypygi is an arachnid order possessing a unique pair of spined pedipalps: appendages that perform in prey capture, courtship, and contest. Pedipalp length, hypothesized to be under sexual selection, varies markedly across amblypygid species, and pedipalp spination, thought to reflect selection for function in prey capture, also differs interspecifically. Differences in pedipalp shape between species may indicate that the relative strength of selection for prey capture and sexual selection vary across the group. However, interspecific differences in pedipalp shape have not been quantified, due to difficulties in identifying homologous features. For the first time, we quantify trends in amblypygid pedipalp shape complexity. We use elliptical Fourier analysis to quantify 2D complexity in pedipalp outlines across eleven species and six genera. We find that complexity significantly decreases as pedipalp length increases. This appears to be driven by relative spine length, suggesting that a trade-off exists between pedipalp length and spination. Furthermore, significant female-biased sexual dimorphism in shape complexity is present in the tibial segment of the amblypygid pedipalp. Our results provide novel insights into the drivers of amblypygid pedipalp evolution and suggest that a functional trade-off between performance in prey capture and other functions under sexual selection exist in this enigmatic structure.

Published
2021

8. A data-based mathematical modelling study to quantify the effects of ciprofloxacin and ampicillin on the within-host dynamics of Salmonella enterica during treatment and relapse

Author

Vlazaki, Myrto, Rossi, Omar, Price, David J, McLean, Callum, Grant, Andrew J, Mastroeni, Pietro, Restif, Olivier, Grant, Andrew [0000-0001-9746-2989], Mastroeni, Pietro [0000-0003-3838-4962], Restif, Olivier [0000-0001-9158-853X], and Apollo - University of Cambridge Repository

Subjects
model selection, data-based model, mechanistic model, isogenic tagged strains, Salmonella enterica, bacterial persistence, Microbial Sensitivity Tests, Models, Theoretical, Anti-Bacterial Agents, Mice, Ciprofloxacin, Recurrence, Animals, Humans, Ampicillin, heterogeneity
Abstract

Antibiotic therapy has drastically reduced the mortality and sequelae of bacterial infections. From naturally occurring to chemically synthesized, different classes of antibiotics have been successfully used without detailed knowledge of how they affect bacterial dynamics in vivo. However, a proportion of patients receiving antimicrobial therapy develop recrudescent infections post-treatment. Relapsing infections are attributable to incomplete clearance of bacterial populations following antibiotic administration; the metabolic profile of this antibiotic-recalcitrant bacterial subpopulation, the spatio-temporal context of its emergence and the variance of antibiotic-bacterial interactions in vivo remain unclear. Here, we develop and apply a mechanistic mathematical model to data from a study comparing the effects of ciprofloxacin and ampicillin on the within-host dynamics of Salmonella enterica serovar Typhimurium in murine infections. Using the inferential capacity of our model, we show that the antibiotic-recalcitrant bacteria following ampicillin, but not ciprofloxacin, treatment belong to a non-replicating phenotype. Aligning with previous studies, we independently estimate that the lymphoid tissues and spleen are important reservoirs of non-replicating bacteria. Finally, we predict that post-treatment, the progenitors of the non-growing and growing bacterial populations replicate and die at different rates. Ultimately, the liver, spleen and mesenteric lymph nodes are all repopulated by progenitors of the previously non-growing phenotype in ampicillin-treated mice.

Published
2020
Full Text
View/download PDF

11. Perceptions of cannabis use: What effect does the perceived use and approval among friends, perceived harm of use, and perceived risk of legal consequences have on the intentions to use cannabis?

Author

Turner-McLean, Callum and Turner-McLean, Callum

Abstract

Cannabis use among young adults is higher than any other age demographic, and research shows that cannabis use increases the risk for adverse mental health, and more so with younger initiation of use. Perceptions of use and approval among friends, and perceptions of harm from use have all been found to predict use. A gap in the research was identified with regards to the effect of the perceived risk of legal consequences on the intentions to use. The present study sought to examine the effect of the perceived risk of arrest and criminal sanctions on the intentions to use cannabis in future and compare this effect with perceived use and approval among friends and perceived harm of cannabis use. A total of 70 young adults aged between 18 and 24 years old completed a questionnaire measuring perceived use and approval among friends, perceived harm of cannabis use, perceived risk of legal consequences, and intentions to use cannabis in the next twelve months. Use and approval among friends, and perceived harm were found to be predictive of intentions to use, perceived risk of legal consequences was not. A multiple regression analysis demonstrated that perceived use and harm were significant predictors, but not approval.

Published
2019

12. Sexual dimorphism in the Arachnid orders.

Author

McLean, Callum J, Garwood, Russell J, Brassey, Charlotte A, McLean, Callum J, Garwood, Russell J, and Brassey, Charlotte A

Abstract

Sexual differences in size and shape are common across the animal kingdom. The study of sexual dimorphism (SD) can provide insight into the sexual- and natural-selection pressures experienced by males and females in different species. Arachnids are diverse, comprising over 100,000 species, and exhibit some of the more extreme forms of SD in the animal kingdom, with the males and females of some species differing dramatically in body shape and/or size. Despite this, research on arachnid SD has primarily focused on specific clades as opposed to observing traits across arachnid orders, the smallest of which have received comparatively little attention. This review provides an overview of the research to date on the trends and potential evolutionary drivers for SD and sexual size dimorphism (SSD) in individual arachnid orders, and across arachnids as a whole. The most common trends across Arachnida are female-biased SSD in total body size, male-biased SSD in relative leg length and SD in pedipalp length and shape. However, the evolution of sexually dimorphic traits within the group is difficult to elucidate due to uncertainty in arachnid phylogenetic relationships. Based on the dataset we have gathered here, we highlight gaps in our current understanding and suggest areas for future research.

Published
2018

14. The kinematics of amblypygid (Arachnida) pedipalps during predation: extreme elongation in raptorial appendages does not result in a proportionate increase in reach and closing speed

Author

McLean, Callum J., Brassey, Charlotte A., Seiter, Michael, Garwood, Russell J., and Gardiner, James D.

Abstract

The link between form and function is key to understanding the evolution of unique and/or extreme morphologies. Amblypygids, or whip spiders, are arachnids that often have highly elongated spined pedipalps. These limbs are used to strike at, and secure, prey before processing by the chelicerae. Amblypygi pedipalps are multifunctional, however, being used in courtship and contest, and vary greatly in form between species. Increased pedipalp length may improve performance during prey capture, but length could also be influenced by factors including territorial contest and sexual selection. Here, for the first time, we used high-speed videography and manual tracking to investigate kinematic differences in prey capture between amblypygid species. Across six morphologically diverse species, spanning four genera and two families, we created a total dataset of 86 trials (9–20 per species). Prey capture kinematics varied considerably between species, with differences being expressed in pedipalp joint angle ranges. In particular, maximum reach ratio did not remain constant with total pedipalp length, as geometric scaling would predict, but decreased with longer pedipalps. This suggests that taxa with the most elongated pedipalps do not deploy their potential length advantage to proportionally increase reach. Therefore, a simple mechanical explanation of increased reach does not sufficiently explain pedipalp elongation. We propose other factors to help explain this phenomenon, such as social interactions or sexual selection, which would produce an evolutionary trade-off in pedipalp length between prey capture performance and other behavioural and/or anatomical pressures.

Published
2024
Full Text
View/download PDF

16. A data-based mathematical modelling study to quantify the effects of ciprofloxacin and ampicillin on the within-host dynamics of Salmonella entericaduring treatment and relapse

Author

Vlazaki, Myrto, Rossi, Omar, Price, David J., McLean, Callum, Grant, Andrew J., Mastroeni, Pietro, and Restif, Olivier

Abstract

Antibiotic therapy has drastically reduced the mortality and sequelae of bacterial infections. From naturally occurring to chemically synthesized, different classes of antibiotics have been successfully used without detailed knowledge of how they affect bacterial dynamics in vivo. However, a proportion of patients receiving antimicrobial therapy develop recrudescent infections post-treatment. Relapsing infections are attributable to incomplete clearance of bacterial populations following antibiotic administration; the metabolic profile of this antibiotic-recalcitrant bacterial subpopulation, the spatio-temporal context of its emergence and the variance of antibiotic–bacterial interactions in vivoremain unclear. Here, we develop and apply a mechanistic mathematical model to data from a study comparing the effects of ciprofloxacin and ampicillin on the within-host dynamics of Salmonella entericaserovar Typhimurium in murine infections. Using the inferential capacity of our model, we show that the antibiotic-recalcitrant bacteria following ampicillin, but not ciprofloxacin, treatment belong to a non-replicating phenotype. Aligning with previous studies, we independently estimate that the lymphoid tissues and spleen are important reservoirs of non-replicating bacteria. Finally, we predict that post-treatment, the progenitors of the non-growing and growing bacterial populations replicate and die at different rates. Ultimately, the liver, spleen and mesenteric lymph nodes are all repopulated by progenitors of the previously non-growing phenotype in ampicillin-treated mice.

Published
2020
Full Text
View/download PDF

Catalog

Books, media, physical & digital resources
See catalog results