Does developmental plasticity influence speciation?

Lead Research Organisation: University of Southampton
Department Name: School of Ocean and Earth Science

Abstract

Life is a journey. As we grow older, we change. Sometimes we respond in the spur of the moment. Occasionally, an event has long-lasting consequences in spite of any change in circumstance and shapes our outlook far into the future.

This future flexibility, or a lack thereof, also applies to the traits like size and weight that influence our daily risk of death and our reproductive success. Some of these traits retain flexibility throughout life, whereas others can only change in a fixed early window. As humans, we are far more likely to shift weight gain trajectories before six months of age than when older.

Any ability to flexibly adjust traits can boost survival chances in new or changing environments, but also provides the means to innovate and so express new combinations of traits. Flexibility as a means of innovation might promote the divergence of ancestral organisms into new species, but also might not because such flexibility would mean that species can already deal with whatever circumstances they encounter, which would in turn remove the pressure for any innovation to become hardwired into their DNA.

The long timescales over which this hardwiring plays out complicates collection of data. We don't know whether future flexibility or a lack of it is more likely to catalyse change into new species. In this project, we will contribute this increasingly requested data and therefore provide the first evidence if a lifetime of flexibility, or a stubborn refusal to change, influences the emergence of new species.

Planktonic foraminifera are single-celled organisms that live in vast numbers in all the world's oceans. While chemical analysis of their fossil remains has generated a remarkably continuous record of past climate change, each individual also retains a complete record of its size and shape at each stage along its journey through life.

These growth stages can be revealed by state-of-the-art imaging technology, which has sparked a digital revolution in how biologists study life on Earth. To study evolution, we need to study differences among lots of individuals. We need to know how and why these differences change through time. This need to measure lots of individuals means that the current practise of a person pointing and clicking on a computer screen to identify distinct parts is too slow. Computer programmes that provide a faster, more repeatable and less biased way of identifying and analysing such parts are now available, completing the toolkit needed to build big databases.

By bringing together lessons from diverse scientific disciplines, we propose to use the same fossil specimens to collate records of an individual's journey through life and the environment it experienced every step of the way, both of which were changing from day-to-day, millions of years ago.

While the fossil record of planktonic foraminifera provides the necessary timespan and abundance, new computer programmes and imaging technology complete the toolkit jigsaw to investigate for the first time if certain parts of an individual's journey through life are more influential than others in determining the eventual evolutionary destinations of its species.

Our unique, direct link between organism and environment lets us study the dynamic journey through life in the static death of the fossil record. The fundamental limitation to the current ways we study how new species emerge is the lack of repeated samples through time to follow the genesis of novel lifeforms, and explicitly targeting this limitation using state-of-the-art approaches from multiple scientific disciplines means we will deliver a breakthrough in attempts to answer one of the most fundamental of all biological questions: how do differences among individuals make differences among species?

Planned Impact

PISTON takes an unprecedented transdisciplinary approach to investigate how new species form. Species are the foundation of biodiversity. The House of Lords Systematics & Taxonomy review identified biodiversity's critical underpinning of an extensive array of natural environment research ranging from blue skies evolutionary questions, particularly into charismatic "missing links" that reveal how differences among individuals generate differences among species, to applied questions such as ecosystem service provision (i.e., the benefits we derive from natural ecosystems) and global health threat mitigation.

PISTON has been designed as a set of interrelated work packages that will integrate to more than the sum of their individual parts to impact policy, public health, industry and the general public. PISTON investigators are active in all these sectors and will build on and further develop existing relationships and activities.

Members of PISTON have a strong track record of contributing to Intergovernmental Panel on Climate Change (IPCC) reports. Our unprecedented data, focusing on the pivotal link between organism and environment and from the last interval when the world was 2-3 degrees warmer than it is today, can contribute to future IPCC reports in this area. Foster (Contributing Author) and Schmidt (Chapter Lead Author; Co-author of Summary for Policymakers) will co-ordinate impact in this area.

Proprietary versions of WP1 computer vision algorithms have successfully delineated the ball and socket of hip joints as a diagnostic to improve hip replacement therapy, which indicates their potential for wider healthcare benefits for the general public. The PISTON team has considerable experience working with industry (Sinclair with Nikon to develop bespoke biological and medical tools) and the public sector (Nixon on, e.g., automated gait recognition) and will draw on this experience to ensure the open-source PISTON outputs are carefully described, annotated and made accessible in toolbox form to the broadest possible community of end-users.

Taxonomic revisions through refined dating and age control will directly impact foraminiferal specialists in commercial operations, particularly the oil industry. The statistical approaches we propose minimise subjective choices, which adds repeatability and transparency across end-users. We will build on existing networks: Wade (NE/N017900) and Wilson (PI: NE/K007211 and NE/K014137) name PetroStrat, Network Stratigraphic, Shell, RPS Energy and Neftex-Halliburton as Industrial Partners on current awards. Wade will act as Industry Impact Champion (1.5% costed time) to incorporate PISTON data and curate the Mikrotax illustrated online portal. In addition to working with External Advisory Board member Haydon Bailey (Network Stratigraphic) to deliver Industrial impact, we will invite further industrial partners to the second scientific scoping workshop in Year 3 (£2500 budgeted for one UK and one international).

All Research Staff will participate in Public Engagement activities during PISTON in areas of particular appeal to each individual. In particular, we will work with Ellen Dowell, a Science Communicator, interdisciplinary facilitator and current collaborator of PI Ezard (NE/J018163), to run interactive workshops in Einstein's Garden at the Green Man Festival in Year 3 (led by Aze), which we will then run subsequently at exhibitions through applications to the NERC Science Festival, Royal Society Summer School and local science festivals in our regions. These hands-on workshops will let festival-goers see evolution in action and contribute to an animated "flick book" style-film based on the divergence into new species (£3000 budgeted). This film will ensure legacy and impact beyond the festival fence.

Studentship Projects

Project Reference Relationship Related To Start End Student Name
NE/P019269/1 01/10/2017 30/09/2023
2383703 Studentship NE/P019269/1 01/03/2018 31/08/2021 Laura Mears
2570968 Studentship NE/P019269/1 29/09/2021 30/03/2025 James Mulqueeney
 
Description PISTON is a transdisciplinary proposal addressing one of the most fundamental of all biological questions: how does variation among individuals generate variation among species? Reviewers described it as "truly ground-breaking in its detail, quality and ambition". Within UoS, I have built new collaborations among Engineers, Computer Scientists, Geochemists and Biologists. We have developed high-throughput repeatable protocols on Synchrotron-equivalent volumetric micro-tomography and laser ablation inductively coupled plasma mass spectrometry. The datasets we have built are already an order of magnitude larger than anything else in the world and ready for analysis into PISTON's scientific questions. My research vision remains to integrate interdisciplinary academic excellence wherever it may be.

As for many researchers, the most recent submission period has presented severe challenges. As PI, I am Clinically Extremely Vulnerable and so have lent on my team to continue delivering the integrated datasets - I have not been inside the building for work for almost a year, but continue to oversee our work packages remotely. The most obvious example of disruption is that we were running stable isotope samples in Ocean & Earth Science's (OES) Organic Geochemistry Facility, part of the SEAPORT Carbon Laboratories, when we went into lockdown 1 and that batch of samples were completed in October 2020. Furthermore, the original OES ICP-MS system has been upgraded to a Triple Quad ICP-MS requiring substantial remote troubleshooting over the autumn. I am happy to report that both facilities are now running again. In particular, the new Triple Quad ICP-MS offers a significant sensitivity improvement and a high degree of software integration with the laser ablation platform. This has streamlined the data acquisition procedure, idea for the large number of analyses required for this project. The initial data was received by the project in late February and endorses the upgrade from the new equipment. Around 75% of our samples are through the first processing stage (x-ray CT) towards the unique integrated datasets we seek to generate, somewhat behind the pre-COVID schedule but not insurmountably given the issues described in this paragraph. I will likely request a short no-cost extension in due course to ensure we hit our initial deliverables.

The major focus across the submission period has been working on the computer algorithms. We successfully recruited a Machine Learning PDRF in December 2020, who has catalysed further development on the computing side of the project; the micropaleontological PDRF has developed the alpha version of her R package to analyse growth trajectories of these foraminifera to push to Github; the RA has completed the manual marking of the "groundtruth" to compare the machine learnt outputs to the human derived ones. Algorithm performance (2/3 of the way through the computer vision work package) is encouraging - the algorithms execute for all ~2700 scans and the returned mean estimate of the number of developmental stages matches the human figure (both 18). Directly comparing the human "groundtruth" and machine learnt outcomes shows that we cannot reject the null hypothesis that both are providing equivalent estimates on an individual specimen level (F, (1, 25) = 0.9333, p>0.05). Nevertheless the variance explained through this relationship is relatively low - this means that while we have an unbiased predictor at present, we would like it to be more precise and continue to work towards driving up the amount of variance explained. We are optimistic of prospects here - most often, the algorithms over-estimate the number of developmental stages and most often that is because the algorithms are detecting juvenile specimens trapped within the focal one. The juveniles have the same repeated structure as the adults, but we need an additional segmentation process to isolate the focal specimen more readily.

We have more publications in later rounds of peer review and anticipate growing levels of outputs as our first fully integrated morphological and geochemical datasets come on stream in the next months.
Exploitation Route Software is currently being developed under proprietary licenses (MATLAB), but is planned to be converted to fully reproducible platforms later. We are settled on an optimal set of statistical outputs from the feature extraction algorithms, all of which will be released in a holistic view of open science. The appointment in December 2020 of PDRF Shetta to the remainder of the role initially filled by PDRF Zhang has catalysed this process. PDRF Brombacher used the period in lockdown to develop the initial version of a package for the R Environment for Statistical and Graphical Computing that is in the very early stages of pushing to github.
Sectors Digital/Communication/Information Technologies (including Software),Education,Culture, Heritage, Museums and Collections,Security and Diplomacy

URL https://www.southampton.ac.uk/oes/research/projects/piston-does-developmental-plasticity-influence-speciation.page
 
Description In very early stages of development work with De La Rue for use of high-resolution computed tomographic images to be used in passport and banknote security features. Julian Payne (Creative Director), Andy Sharp (Photographer) and Stuart Rost (Designer) from De La Rue visited in September 2018 to discuss plans, capture images and talk to team members. Unclear of future use at this stage (follow-up meeting in April 2019). Images will become part of a National Trust exhibition on digital nature from May 2019; precise involvement to be clarified and led by University of Bristol component of the award. In 2019, working with colleagues in the University of Southampton's IT department, we developed a fully interactive Virtual Reality experience "climbing" inside the shells of these plankton to crawl back through their life histories to the earliest stages of development that are some 15 um in diameter. We have beta-tested the experience with 3rd year undergraduate students within the Microfossils, Environments, Evolution and Time module at the University of Southampton and will roll out related public engagement activities based on these experiences in 2020. Ezard is now a member of the University-wide Virtual Reality Special Interest Group co-ordinating the best ways to develop such features across a Research, Education and Enterprise portfolio.
First Year Of Impact 2018
Sector Digital/Communication/Information Technologies (including Software),Education,Culture, Heritage, Museums and Collections,Security and Diplomacy
Impact Types Cultural,Societal

 
Description Improving Digital Security features on passports and banknotes
Geographic Reach Local/Municipal/Regional 
Policy Influence Type Influenced training of practitioners or researchers
Impact The x-ray CT images we have developed were used by digital security firm De La Rue to update their digital security features. As we have built the largest collection of intraspecific variation x-ray CT scans (by an order of magnitude), we have many similar images with fine scale variation in micro-morphological features. This natural variation in the porosity of plankton makes bank notes and passports exceptionally difficult to forge when these images are used as backdrops. De La Rue Creative Director Julian Payne wrote "without your involvement we would not have focussed on the University of Southampton." The final prototypes used medical imagery rather than our plankton for ease of communication with stakeholders (i.e. De La Rue customers), but we were a crucial stage in the Research and Development phase.
 
Description Palaeoenvironmental controls on dispersal of planktic foraminifera in the Plio-Pleistocene oceans
Amount £50,000 (GBP)
Organisation Natural Environment Research Council 
Sector Public
Country United Kingdom
Start 01/2020 
End 06/2020
 
Description The contribution of plasticity to adaptive divergence: domestication as a model
Amount £473,142 (GBP)
Funding ID NE/S002022/1 
Organisation Natural Environment Research Council 
Sector Public
Country United Kingdom
Start 01/2019 
End 12/2021