Limits...
Evolving nutritional strategies in the presence of competition: a geometric agent-based model.

Senior AM, Charleston MA, Lihoreau M, Buhl J, Raubenheimer D, Simpson SJ - PLoS Comput. Biol. (2015)

Bottom Line: Our results indicate that a transition between two alternative strategies occurs at moderate to high levels of competition.Our models imply that competition combined with reproductive skew in social groups can play a role in the evolution of diet breadth.We discuss how the integration of agent-based, nutritional and evolutionary modelling may be applied in future studies to further understand the evolution of nutritional strategies across social and ecological contexts.

View Article: PubMed Central - PubMed

Affiliation: Charles Perkins Centre, The University of Sydney, Sydney, New South Wales, Australia; School of Biological Sciences, The University of Sydney, Sydney, New South Wales, Australia.

ABSTRACT
Access to nutrients is a key factor governing development, reproduction and ultimately fitness. Within social groups, contest-competition can fundamentally affect nutrient access, potentially leading to reproductive asymmetry among individuals. Previously, agent-based models have been combined with the Geometric Framework of nutrition to provide insight into how nutrition and social interactions affect one another. Here, we expand this modelling approach by incorporating evolutionary algorithms to explore how contest-competition over nutrient acquisition might affect the evolution of animal nutritional strategies. Specifically, we model tolerance of nutrient excesses and deficits when ingesting nutritionally imbalanced foods, which we term 'nutritional latitude'; a higher degree of nutritional latitude constitutes a higher tolerance of nutritional excess and deficit. Our results indicate that a transition between two alternative strategies occurs at moderate to high levels of competition. When competition is low, individuals display a low level of nutritional latitude and regularly switch foods in search of an optimum. When food is scarce and contest-competition is intense, high nutritional latitude appears optimal, and individuals continue to consume an imbalanced food for longer periods before attempting to switch to an alternative. However, the relative balance of nutrients within available foods also strongly influences at what levels of competition, if any, transitions between these two strategies occur. Our models imply that competition combined with reproductive skew in social groups can play a role in the evolution of diet breadth. We discuss how the integration of agent-based, nutritional and evolutionary modelling may be applied in future studies to further understand the evolution of nutritional strategies across social and ecological contexts.

No MeSH data available.


The Models Implementation of the Geometric Framework (GF).An example of our implementation of the GF, redrawn from Lihoreau et al. [5]. The x and y axes represent protein (P) and carbohydrate (C). The intake target (IT) is denoted by the red crosshair and the individual’s current nutritional state by the black point. The food rail for the food an individual is consuming is given by the black line (f) with the angle αf. The amount of food an individual would eat to maximise its fitness is given by the individual’s appetite (A). A is governed by the nearest distance rule of compromise; an individual gets as close to the IT as the food rail allows [21]. A is the scalar projection of the Euclidean distance between an individual’s nutritional state and the IT on to the food rail f. A is found by estimating the ‘ideal’ food rail that connects the individual’s nutritional state with the IT (dashed line with angle αideal), the magnitude of the vector along which an individual would travel to reach the IT (//VT//) and the angle between αf and αideal (β); Equation 4. Note that the amount an individual can eat in one time step has a maximum value of φ.
© Copyright Policy
Related In: Results  -  Collection

License
getmorefigures.php?uid=PMC4376532&req=5

pcbi.1004111.g009: The Models Implementation of the Geometric Framework (GF).An example of our implementation of the GF, redrawn from Lihoreau et al. [5]. The x and y axes represent protein (P) and carbohydrate (C). The intake target (IT) is denoted by the red crosshair and the individual’s current nutritional state by the black point. The food rail for the food an individual is consuming is given by the black line (f) with the angle αf. The amount of food an individual would eat to maximise its fitness is given by the individual’s appetite (A). A is governed by the nearest distance rule of compromise; an individual gets as close to the IT as the food rail allows [21]. A is the scalar projection of the Euclidean distance between an individual’s nutritional state and the IT on to the food rail f. A is found by estimating the ‘ideal’ food rail that connects the individual’s nutritional state with the IT (dashed line with angle αideal), the magnitude of the vector along which an individual would travel to reach the IT (//VT//) and the angle between αf and αideal (β); Equation 4. Note that the amount an individual can eat in one time step has a maximum value of φ.

Mentions: Eat: The ‘Eat’ process is summarised in Fig. 9. Eating consists of moving through the nutrient space at an angle (or heading) given by the food rail for the food on which the individual is eating; αf. The distance an individual moves (i.e., the amount eaten) is governed by the individual’s appetite (A), which is assumed to conform to the ‘nearest distance’ rule of compromise [23] and the maximum amount of food an individual can eat (φ) following Equation 4:distancemoved=min{A,φ},(4)where A is calculated as Equation 5:A=‖Vt‖cosβ,(5)where β is the angle between the food rail on which the animal is feeding (αf) and the ‘ideal’ food rail that joins the individual’s current nutritional state with the IT (αideal) and VT is the vector connecting the individual’s current nutritional state and the IT; A thus gives the distance the individual would move to reach the point of nutritional compromise (Fig. 9). If this distance is greater than the maximum amount the individual can eat in one time-step (φ) then the individual’s nutritional state moves by φ.


Evolving nutritional strategies in the presence of competition: a geometric agent-based model.

Senior AM, Charleston MA, Lihoreau M, Buhl J, Raubenheimer D, Simpson SJ - PLoS Comput. Biol. (2015)

The Models Implementation of the Geometric Framework (GF).An example of our implementation of the GF, redrawn from Lihoreau et al. [5]. The x and y axes represent protein (P) and carbohydrate (C). The intake target (IT) is denoted by the red crosshair and the individual’s current nutritional state by the black point. The food rail for the food an individual is consuming is given by the black line (f) with the angle αf. The amount of food an individual would eat to maximise its fitness is given by the individual’s appetite (A). A is governed by the nearest distance rule of compromise; an individual gets as close to the IT as the food rail allows [21]. A is the scalar projection of the Euclidean distance between an individual’s nutritional state and the IT on to the food rail f. A is found by estimating the ‘ideal’ food rail that connects the individual’s nutritional state with the IT (dashed line with angle αideal), the magnitude of the vector along which an individual would travel to reach the IT (//VT//) and the angle between αf and αideal (β); Equation 4. Note that the amount an individual can eat in one time step has a maximum value of φ.
© Copyright Policy
Related In: Results  -  Collection

License
Show All Figures
getmorefigures.php?uid=PMC4376532&req=5

pcbi.1004111.g009: The Models Implementation of the Geometric Framework (GF).An example of our implementation of the GF, redrawn from Lihoreau et al. [5]. The x and y axes represent protein (P) and carbohydrate (C). The intake target (IT) is denoted by the red crosshair and the individual’s current nutritional state by the black point. The food rail for the food an individual is consuming is given by the black line (f) with the angle αf. The amount of food an individual would eat to maximise its fitness is given by the individual’s appetite (A). A is governed by the nearest distance rule of compromise; an individual gets as close to the IT as the food rail allows [21]. A is the scalar projection of the Euclidean distance between an individual’s nutritional state and the IT on to the food rail f. A is found by estimating the ‘ideal’ food rail that connects the individual’s nutritional state with the IT (dashed line with angle αideal), the magnitude of the vector along which an individual would travel to reach the IT (//VT//) and the angle between αf and αideal (β); Equation 4. Note that the amount an individual can eat in one time step has a maximum value of φ.
Mentions: Eat: The ‘Eat’ process is summarised in Fig. 9. Eating consists of moving through the nutrient space at an angle (or heading) given by the food rail for the food on which the individual is eating; αf. The distance an individual moves (i.e., the amount eaten) is governed by the individual’s appetite (A), which is assumed to conform to the ‘nearest distance’ rule of compromise [23] and the maximum amount of food an individual can eat (φ) following Equation 4:distancemoved=min{A,φ},(4)where A is calculated as Equation 5:A=‖Vt‖cosβ,(5)where β is the angle between the food rail on which the animal is feeding (αf) and the ‘ideal’ food rail that joins the individual’s current nutritional state with the IT (αideal) and VT is the vector connecting the individual’s current nutritional state and the IT; A thus gives the distance the individual would move to reach the point of nutritional compromise (Fig. 9). If this distance is greater than the maximum amount the individual can eat in one time-step (φ) then the individual’s nutritional state moves by φ.

Bottom Line: Our results indicate that a transition between two alternative strategies occurs at moderate to high levels of competition.Our models imply that competition combined with reproductive skew in social groups can play a role in the evolution of diet breadth.We discuss how the integration of agent-based, nutritional and evolutionary modelling may be applied in future studies to further understand the evolution of nutritional strategies across social and ecological contexts.

View Article: PubMed Central - PubMed

Affiliation: Charles Perkins Centre, The University of Sydney, Sydney, New South Wales, Australia; School of Biological Sciences, The University of Sydney, Sydney, New South Wales, Australia.

ABSTRACT
Access to nutrients is a key factor governing development, reproduction and ultimately fitness. Within social groups, contest-competition can fundamentally affect nutrient access, potentially leading to reproductive asymmetry among individuals. Previously, agent-based models have been combined with the Geometric Framework of nutrition to provide insight into how nutrition and social interactions affect one another. Here, we expand this modelling approach by incorporating evolutionary algorithms to explore how contest-competition over nutrient acquisition might affect the evolution of animal nutritional strategies. Specifically, we model tolerance of nutrient excesses and deficits when ingesting nutritionally imbalanced foods, which we term 'nutritional latitude'; a higher degree of nutritional latitude constitutes a higher tolerance of nutritional excess and deficit. Our results indicate that a transition between two alternative strategies occurs at moderate to high levels of competition. When competition is low, individuals display a low level of nutritional latitude and regularly switch foods in search of an optimum. When food is scarce and contest-competition is intense, high nutritional latitude appears optimal, and individuals continue to consume an imbalanced food for longer periods before attempting to switch to an alternative. However, the relative balance of nutrients within available foods also strongly influences at what levels of competition, if any, transitions between these two strategies occur. Our models imply that competition combined with reproductive skew in social groups can play a role in the evolution of diet breadth. We discuss how the integration of agent-based, nutritional and evolutionary modelling may be applied in future studies to further understand the evolution of nutritional strategies across social and ecological contexts.

No MeSH data available.