Machine learning to detect marine animals in UAV imagery: effect of morphology, spacing, behaviour and habitat

Authors: Antoine M. Dujon; Daniel Ierodiaconou; Johanna J. Geeson; John P. Y. Arnould; Blake M. Allan; Kostas A. Katselidis; Gail Schofield

Source: Remote Sensing in Ecology & Conservation (MAY 2020)

Brief summary of the paper:

Machine learning algorithms are being increasingly used to process large volumes of wildlife imagery data from unmanned aerial vehicles (UAVs); however, suitable algorithms to monitor multiple species are required to enhance efficiency.

Here, we developed a machine learning algorithm using a low‐cost computer. We trained a convolutional neural network and tested its performance in: (1) distinguishing focal organisms of three marine taxa (Australian fur seals, loggerhead sea turtles and Australasian gannets; body size ranges: 0.8–2.5 m, 0.6–1.0 m, and 0.8–0.9 m, respectively); and (2) simultaneously delineating the fine‐scale movement trajectories of multiple sea turtles at a fish cleaning station.

For all species, the algorithm performed best at detecting individuals of similar body length, displaying consistent behaviour or occupying uniform habitat (proportion of individuals detected, or recall of 0.94, 0.79 and 0.75 for gannets, seals and turtles, respectively). For gannets, performance was impacted by spacing (huddling pairs with offspring) and behaviour (resting vs. flying shapes, overall precision: 0.74).

For seals, accuracy was impacted by morphology (sexual dimorphism and pups), spacing (huddling and creches) and habitat complexity (seal sized boulders) (overall precision: 0.27). For sea turtles, performance was impacted by habitat complexity, position in water column, spacing, behaviour (interacting individuals) and turbidity (overall precision: 0.24); body size variation had no impact.

For sea turtle trajectories, locations were estimated with a relative positioning error of <50 cm. In conclusion, we demonstrate that, while the same machine learning algorithm can be used to survey multiple species, no single algorithm captures all components optimally within a given site.

We recommend that, rather than attempting to fully automate detection of UAV imagery data, semi‐automation is implemented (i.e. part automated and part manual, as commonly practised for photo‐identification). Approaches to enhance the efficiency of manual detection are required in parallel to the development of effective implementation of machine learning algorithms.

Linking pollution and cancer in aquatic environments: A review

Authors: Ciara Baines; Adelaide Lerebours; Frederic Thomas; Jerome Fort; Randel Kreitsberg; Sophie Gentes; Richard Meitern; Lauri Saks; Beata Ujvari; Mathieu Giraudeau; Tuul Seppa


Brief summary of the paper:

Due to the interconnectedness of aquatic ecosystems through the highly effective marine and atmospheric transport routes, all aquatic ecosystems are potentially vulnerable to pollution. Whilst links between pollution and increased mortality of wild animals have now been firmly established, the next steps should be to focus on specific physiological pathways and pathologies that link pollution to wildlife health deterioration.

One of the pollution-induced pathologies that should be at the centre of attention in ecological and evolutionary research is cancer, as anthropogenic contamination has resulted in a rapid increase of oncogenic substances in natural habitats. Whilst wildlife cancer research is an emerging research topic, systematic reviews of the many case studies published over the recent decades are scarce.

This research direction would (1) provide a better understanding of the physiological mechanisms connecting anthropogenic pollution to oncogenic processes in non-model organisms (reducing the current bias towards human and lab-animal studies in cancer research), and (2) allow us to better predict the vulnerability of different wild populations to oncogenic contamination. This article combines the information available within the scientific literature about cancer occurrences in aquatic and semi-aquatic species.

For the first aim, we use available knowledge from aquatic species to suggest physiological mechanisms that link pollution and cancer, including main metabolic detoxification pathways, oxidative damage effects, infections, and changes to the microbiome.

For the second aim, we determine which types of aquatic animals are more vulnerable to pollution-induced cancer, which types of pollution are mainly associated with cancer in aquatic ecosystems, and which types of cancer pollution causes.

We also discuss the role of migration in exposing aquatic and semi-aquatic animals to different oncogenic pollutants. Finally, we suggest novel research avenues, including experimental approaches, analysis of the effects of pollutant cocktails and long-term chronic exposure to lower levels of pollutants, and the use of already published databases of gene expression levels in animals from differently polluted habitats.

Identifying key questions in the ecology and evolution of cancer

Authors: Antoine M. Dujon; Athena Aktipis; Catherine Alix‐Panabières; Sarah R. Amend; Amy M. Boddy; Joel S. Brown; Jean‐Pascal Capp; James DeGregori; Paul Ewald; Robert Gatenby; Marco Gerlinger; Mathieu Giraudeau; Rodrigo K. Hamede; Elsa Hansen; Irina Kareva; Carlo C. Maley; Andriy Marusyk; Nicholas McGranahan; Michael J. Metzger; Aurora M. Nedelcu; Robert Noble; Leonard Nunney; Kenneth J. Pienta; Kornelia Polyak; Pascal Pujol; Andrew F. Read; Benjamin Roche; Susanne Sebens; Eric Solary; Kateřina Staňková; Holly Swain Ewald; Frédéric Thomas & Beata Ujvari

Source: Evolutionary Applications (DEC 2020)

Brief summary of the paper:

The application of evolutionary and ecological principles to cancer prevention and treatment, as well as recognising cancer as a selection force in nature, has gained impetus over the last 50 years.

Following the initial theoretical approaches that combined knowledge from interdisciplinary fields, it became clear that using the eco‐evolutionary framework is of key importance to understand cancer. We are now at a pivotal point where accumulating evidence start to steer the future directions of the discipline and allow us to underpin the key challenges that remains to be addressed.

Here, we aim to assess current advancements in the field, and to suggest future directions for research. First, we summarize cancer research areas that, so far, have assimilated ecological and evolutionary principles into their approaches and illustrate of their key importance.

Then, we assembled 33 experts and identified 84 key questions, organized around nine major themes, to pave the foundations for research to come. We highlight the urgent need for broadening the portfolio of research directions to stimulate novel approaches at the interface of oncology and ecological and evolutionary sciences.

We conclude that progressive and efficient cross‐disciplinary collaborations that draw on the expertise of the fields of ecology, evolution and cancer are essential in order to efficiently address current and future questions about cancer.

Transmissible cancers in mammals and bivalves: How many examples are there?

Predictions indicate widespread occurrence

Authors: Antoine M. Dujon; Georgina Bramwell; Benjamin Roche; Frédéric Thomas & Beata Ujvari

Source: BioEssays (NOV 2020)

Brief summary of the paper:

Transmissible cancers are elusive and understudied parasitic life forms caused by malignant clonal cells (nine lineages are known so far). They emerge by completing sequential steps that include breaking cell cooperation, evade anti‐cancer defences and shedding cells to infect new hosts. Transmissible cancers impair host fitness, and their importance as selective force is likely largely underestimated. It is, therefore, crucial to determine how common they might be in the wild.

Here, we draw a parallel between the steps required for a transmissible cancer to emerge and the steps required for an intelligent civilisation to emerge in the Milky Way using a modified Drake equation. Using numerical analyses, we estimate the potential number of extant marine and bivalve species in which transmissible cancers might exist.

Our results suggest that transmissible cancers are more common than expected, and that new lineages can be found by screening a large number of species.

Do malignant cells sleep at night?

Authors: Luis Enrique Cortés-Hernández, Zahra Eslami-S, Antoine M. Dujon, Mathieu Giraudeau, Beata Ujvari, Frédéric Thomas & Catherine Alix-Panabières

Source: Genome Biology (NOV 2020)

Brief summary of the paper:

Biological rhythms regulate the biology of most, if not all living creatures, from whole organisms to their constitutive cells, their microbiota, and also parasites.

Here, we present the hypothesis that internal and external ecological variations induced by biological cycles also influence or are exploited by cancer cells, especially by circulating tumor cells, the key players in the metastatic cascade.

We then discuss the possible clinical implications of the effect of biological cycles on cancer progression, and how they could be exploited to improve and standardize methods used in the liquid biopsy field.

Unravelling the cancer puzzle from an ecological and evolutionary perspective – an Australian and French International Associated Laboratory

In case you’ve missed this, a recording of the conference – “Unravelling the cancer puzzle from an ecological and evolutionary perspective – an Australian and French International Associated Laboratory” – is now available online. Click HERE to watch the recording.

Presenters Professor Frédéric Thomas (CNRS, France) and Associate Professor Beata Ujvari (Deakin University) will focus on how applying evolutionary principles to cancer revolutionises treatment strategies and approaches.

The Melbourne Salon is a joint venture between RMIT University, the Institute for the Study of French-Australian Relations and the Alliance Française de Melbourne. It is a place where curious and open-minded people can engage in French-Australian cross-cultural dialogues with the aim of broadening the understanding of French and Australian perspectives on contemporary topics through conversation, debate and discussion.

Genetic rescue restores long-term viability of an isolated population of adders (Vipera berus)

Authors: Thomas Madsen; Jon Loman; Lewis Anderberg; Håkan Anderberg; Arthur Georges & Beata Ujvari

Source: Current Biology (NOV 2020)

Brief summary of the paper:

Climate change is regarded as a major threat to global biodiversity [1]. However, another key driver of declines in biodiversity during the last century has been, and still is, the devastating impact of anthropogenic habitat destruction.

Human degradation of natural habitats has resulted in large, formerly homogeneous areas becoming exceedingly isolated and fragmented, resulting in reduced genetic diversity and a concomitant increased vulnerability to pathogens and increased risk of inbreeding.

In order to restore genetic diversity in small isolated or fragmented populations, genetic rescue — that is, an intervention in which unrelated individuals are brought into a population, leading to introduction of novel alleles — has been shown to reduce the deleterious effects of inbreeding

Cancer risk landscapes: A framework to study cancer in ecosystems

Authors: Antoine M. Dujon; Beata Ujvari; FrédéricThomas

Source: Science of The Total Environment (OCT 2020)

Brief summary of the paper:

Cancer is a family of diseases that has been documented in most metazoan species and ecosystems. Human induced environmental changes are increasingly exposing wildlife to carcinogenic risk factors, and negative repercussions on ecosystems and on the conservation of endangered species are already been observed.

It is therefore of key importance to understand the spatiotemporal variability of those risk factors and how they interact with the biosphere to mitigate their effects.

Here we introduce the concept of cancer risk landscape that can be applied to understand how species are exposed to, interact with, and modify cancer risk factors.

With this publication we aim to provide a framework in order to stimulate a discussion on how to mitigate cancer-causing risk factors.

Ecological and evolutionary consequences of anticancer adaptations

Authors: Justine BOUTRY; Antoine DUJON; G.E.R.A.R.D. Anne-Lise; Sophie TISSOT; Nick MACDONALD; Aaron SCHULTZ; Peter A. BIRO; Christa BECKMANN; Rodrigo HAMEDE; David G. HAMILTON; Mathieu GIRAUDEAU; Beata UJVARI; Frédéric THOMAS

Source: iScience (Oct 2020)

Brief summary of the paper:

Cellular cheating leading to cancers exists in all branches of multicellular life, favoring the evolution of adaptations to avoid or suppress malignant progression, and/or to alleviate its fitness consequences.

Ecologists have until recently largely neglected the importance of cancer cells for animal ecology, presumably because they did not consider either the potential ecological or evolutionary consequences of anticancer adaptations.

Here, we review the diverse ways in which the evolution of anticancer adaptations has significantly constrained several aspects of the evolutionary ecology of multicellular organisms at the cell, individual, population, species and ecosystem levels, and suggest some avenues for future research.

Long term effects of outbreeding: experimental founding of island population eliminates malformations and improves hatching success in sand lizards

Authors: Willow R. Lindsay; Thomas Madsen; Erik Wapstra; Mette Lillie; Lisa Loeb; Beata Ujvari; Mats Olsson


Brief summary of the paper:

Loss of genetic variation is an increasing problem in many natural populations as a result of population fragmentation, inbreeding, and genetic drift, which may lead to inbreeding depression and subsequent “extinction vortices”. In such cases, outbreeding offers a potential population saviour from extinction.

Here we compare offspring viability between an experimentally founded outbred island population of sand lizards Lacerta agilis, and an inbred mainland source population on the Swedish West coast.

We have studied the mainland population for over a decade during which >4000 offspring from >500 parents were monitored. We conducted an outbreeding experiment in which lizards from the mainland population with relatively low genetic variation were crossbred with lizards from distant populations that lack gene flow.

The resulting 454 offspring were introduced to an otherwise uninhabited island with ideal sand lizard habitat. A survey of the island two decades later showed that offspring produced by females from the experimentally founded population had 13% higher hatching success (99.3% versus 86.4%) and elimination of the malformations occurring in 21% of clutches in the mainland source population.

These results co-occur with higher genetic diversity. We conclude that outbreeding improved offspring viability in our island population ca 5–6 generations after the founding event, that is, with sustained viability effects at a time when heterotic effects are expected to have subsided.