Machine-learning model led design to experimentally test species thermal limits: the case of kissing bugs (Triatominae)

AbstractSpecies Distribution Modelling (SDM) determines habitat suitability of a species across geographic areas using macro-climatic variables; however, micro-habitats can buffer or exacerbate the influence of macro-climatic variables, requiring links between physiology and species persistence. Experimental approaches linking species physiology to micro-climate are complex, time consuming and expensive. E.g., what combination of exposure time and temperature is important for a species thermal tolerance is difficult to judge a priori. We tackled this problem using an active learning approach that utilized machine learning methods to guide thermal tolerance experimental design for three kissing-bug species (Hemiptera: Reduviidae: Triatominae), vectors of the parasite causing Chagas disease. As with other pathogen vectors, triatomines are well known to utilize micro-habitats and the associated shift in microclimate to enhance survival. Using a limited literature-collected dataset, our approach showed that temperature followed by exposure time were the strongest predictors of mortality; species played a minor role, and life stage was the least important. Further, we identified complex but biologically plausible nonlinear interactions between temperature and exposure time in shaping mortality, together setting the potential thermal limits of triatomines. The results from this data led to the design of new experiments with laboratory results that produced novel insights of the effects of temperature and exposure for the triatomines. These results, in turn, can be used to better model micro-climatic envelope for the species. Here we demonstrate the power of an active learning approach to explore experimental space to design laboratory studies testing species thermal limits. Our analytical pipeline can be easily adapted to other systems and we provide code to allow practitioners to perform similar analyses. Not only does our approach have the potential to save time and money: it can also increase our understanding of the links between species physiology and climate, a topic of increasing ecological importance.Author summarySpecies Distribution Modelling determines habitat suitability of a species across geographic areas using macro-climatic variables; however, micro-habitats can buffer or exacerbate the influence of macro-climatic variables, requiring links between physiology and species persistence. We tackled the problem of the combination of exposure time and temperature (a combination difficult to judge a priori) in determining species thermal tolerance, using an active learning approach that utilized machine learning methods to guide thermal tolerance experimental design for three kissing-bug species, vectors of the parasite causing Chagas disease. These bugs are found in micro-habitats with associated shifts in microclimate to enhance survival. Using a limited literature-collected dataset, we showed that temperature followed by exposure time were the strongest predictors of mortality, that species played a minor role, that life stage was the least important, and a complex nonlinear interaction between temperature and exposure time in shaping mortality of kissing bugs. These results led to the design of new laboratory experiments to assess the effects of temperature and exposure for the triatomines. These results can be used to better model micro-climatic envelope for species. Our active learning approach to explore experimental space to design laboratory studies can also be applied to other environmental conditions or species.

Download Full-text

Machine-learning model led design to experimentally test species thermal limits: The case of kissing bugs (Triatominae)

PLoS Neglected Tropical Diseases ◽

10.1371/journal.pntd.0008822 ◽

2021 ◽

Vol 15 (3) ◽

pp. e0008822

Author(s):

Jorge E. Rabinovich ◽

Agustín Alvarez Costa ◽

Ignacio J. Muñoz ◽

Pablo E. Schilman ◽

Nicholas M. Fountain-Jones

Keyword(s):

Machine Learning ◽

Active Learning ◽

Exposure Time ◽

Thermal Tolerance ◽

Life Stage ◽

Climatic Variables ◽

Triatoma Infestans ◽

Learning Approach ◽

Test Species ◽

Thermal Limits

Species Distribution Modelling (SDM) determines habitat suitability of a species across geographic areas using macro-climatic variables; however, micro-habitats can buffer or exacerbate the influence of macro-climatic variables, requiring links between physiology and species persistence. Experimental approaches linking species physiology to micro-climate are complex, time consuming and expensive. E.g., what combination of exposure time and temperature is important for a species thermal tolerance is difficult to judge a priori. We tackled this problem using an active learning approach that utilized machine learning methods to guide thermal tolerance experimental design for three kissing-bug species: Triatoma infestans, Rhodnius prolixus, and Panstrongylus megistus (Hemiptera: Reduviidae: Triatominae), vectors of the parasite causing Chagas disease. As with other pathogen vectors, triatomines are well known to utilize micro-habitats and the associated shift in microclimate to enhance survival. Using a limited literature-collected dataset, our approach showed that temperature followed by exposure time were the strongest predictors of mortality; species played a minor role, and life stage was the least important. Further, we identified complex but biologically plausible nonlinear interactions between temperature and exposure time in shaping mortality, together setting the potential thermal limits of triatomines. The results from this data led to the design of new experiments with laboratory results that produced novel insights of the effects of temperature and exposure for the triatomines. These results, in turn, can be used to better model micro-climatic envelope for the species. Here we demonstrate the power of an active learning approach to explore experimental space to design laboratory studies testing species thermal limits. Our analytical pipeline can be easily adapted to other systems and we provide code to allow practitioners to perform similar analyses. Not only does our approach have the potential to save time and money: it can also increase our understanding of the links between species physiology and climate, a topic of increasing ecological importance.

Download Full-text

An Automated Machine Learning-Genetic Algorithm Framework With Active Learning for Design Optimization

Journal of Energy Resources Technology ◽

10.1115/1.4050489 ◽

2021 ◽

Vol 143 (8) ◽

Author(s):

Opeoluwa Owoyele ◽

Pinaki Pal ◽

Alvaro Vidal Torreira

Keyword(s):

Machine Learning ◽

Active Learning ◽

Design Optimization ◽

Fuel Injection ◽

A Priori ◽

Computational Cost ◽

Learning Approach ◽

Data Generation ◽

Compression Ignition Engine ◽

Hyperparameter Selection

AbstractThe use of machine learning (ML)-based surrogate models is a promising technique to significantly accelerate simulation-driven design optimization of internal combustion (IC) engines, due to the high computational cost of running computational fluid dynamics (CFD) simulations. However, training the ML models requires hyperparameter selection, which is often done using trial-and-error and domain expertise. Another challenge is that the data required to train these models are often unknown a priori. In this work, we present an automated hyperparameter selection technique coupled with an active learning approach to address these challenges. The technique presented in this study involves the use of a Bayesian approach to optimize the hyperparameters of the base learners that make up a super learner model. In addition to performing hyperparameter optimization (HPO), an active learning approach is employed, where the process of data generation using simulations, ML training, and surrogate optimization is performed repeatedly to refine the solution in the vicinity of the predicted optimum. The proposed approach is applied to the optimization of a compression ignition engine with control parameters relating to fuel injection, in-cylinder flow, and thermodynamic conditions. It is demonstrated that by automatically selecting the best values of the hyperparameters, a 1.6% improvement in merit value is obtained, compared to an improvement of 1.0% with default hyperparameters. Overall, the framework introduced in this study reduces the need for technical expertise in training ML models for optimization while also reducing the number of simulations needed for performing surrogate-based design optimization.

Download Full-text

Disparate patterns of thermal adaptation between life stages in temperate vs. tropical Drosophila melanogaster

10.1101/120360 ◽

2017 ◽

Author(s):

Brent L. Lockwood ◽

Tarun Gupta ◽

Rosemary Scavotto

Keyword(s):

Climate Change ◽

Drosophila Melanogaster ◽

Heat Tolerance ◽

Thermal Tolerance ◽

Life Stage ◽

Thermal Adaptation ◽

Life Stages ◽

Thermal Limits ◽

Heat Tolerant ◽

Selection For

AbstractMany terrestrial ectothermic species exhibit limited variation in upper thermal tolerance across latitude. However, these trends may not signify limited adaptive capacity to increase thermal tolerance in the face of climate change. Instead, thermal tolerance may be similar among populations because behavioral thermoregulation by mobile organisms or life stages may buffer natural selection for thermal tolerance. We compared thermal tolerance of adults and embryos among natural populations of Drosophila melanogaster from a broad range of thermal habitats around the globe to assess natural variation of thermal tolerance in mobile vs. immobile life stages. We found no variation among populations in adult thermal tolerance, but embryonic thermal tolerance was higher in tropical strains than in temperate strains. Average maximum temperature of the warmest month of the year predicted embryonic thermal tolerance in tropical but not temperate sites. We further report that embryos live closer to their upper thermal limits than adultso—i.e., thermal safety margins are smaller for embryos than adults. F1 hybrid embryos from crosses between temperate and tropical populations had thermal tolerance that matched that of tropical embryos, suggesting dominance of heat-tolerant alleles. Together our findings suggest that thermal selection has led to divergence in embryonic thermal tolerance but that selection for divergent thermal tolerance may be limited in adults. Further, our results suggest that thermal traits should be measured across life stages in order to better predict adaptive limits.Impact SummaryClimate change may threaten the extinction of many ectothermic species, unless populations can evolutionarily adapt to rising temperatures. Natural selection should favor individuals with higher heat tolerances in hotter environments. But recent studies have found that individuals from hot and cold places often have similar heat tolerances. This pattern may indicate that the evolution of heat tolerance is constrained. If this were true, then it would have dire consequences for species persistence under novel thermal conditions.An alternative explanation for lack of variation in heat tolerance is that mobile organisms don’t need higher heat tolerances to survive in hotter places. The majority of studies have focused on heat tolerance of the adult life stage. Yet, adults in many species are mobile organisms that can avoid extreme heat by seeking shelter in cooler microhabitats (e.g., shaded locations). In contrast, immobile life stages (e.g., insect eggs) cannot behaviorally avoid extreme heat. Thus, mobile and immobile life stages may face different thermal selection pressures that lead to disparate patterns of thermal adaptation across life stages.Here, we compared heat tolerances of fruit fly adults and eggs (Drosophila melanogaster) from populations in temperate North America and tropical locations around the globe. Consistent with previous studies, we found no differences among populations in adult heat tolerance. However, eggs from tropical flies were consistently more heat tolerant than eggs from North American flies. Further, eggs had lower heat tolerance than adults. Consequently, fly eggs in the hotter tropics may experience heat death more frequently than adult flies later in life. This may explain why patterns of divergence in heat tolerance were decoupled across life stages. These patterns indicate that thermal adaptation may be life-stage-specific and suggest that future work should characterize thermal traits across life stages to better understand the evolution of thermal limits.

Download Full-text

An Automated Machine Learning-Genetic Algorithm (AutoML-GA) Framework With Active Learning for Design Optimization

ASME 2020 Internal Combustion Engine Division Fall Technical Conference ◽

10.1115/icef2020-3000 ◽

2020 ◽

Author(s):

Opeoluwa Owoyele ◽

Pinaki Pal ◽

Alvaro Vidal Torreira

Keyword(s):

Machine Learning ◽

Active Learning ◽

Design Optimization ◽

Test Performance ◽

Computational Cost ◽

Learning Approach ◽

Data Generation ◽

Compression Ignition Engine ◽

Ic Engine ◽

Hyperparameter Selection

Abstract The use of machine learning (ML) based surrogate models is a promising technique to significantly accelerate simulation-based design optimization of IC engines, due to the high computational cost of running computational fluid dynamics (CFD) simulations. However, surrogate-based optimization for IC engine applications suffers from two main issues. First, training ML models requires hyperparameter selection, often involving trial-and-error combined with domain expertise. The second issue is that the data required to train these models is often unknown a priori. In this work, we present an automated hyperparameter selection technique coupled with an active learning approach to address these challenges. The technique presented in this study involves the use of a Bayesian approach to optimize the hyperparameters of the base learners that make up a Super Learner model to obtain better test performance. In addition to performing hyperparameter optimization (HPO), an active learning approach is employed, where the process of data generation using simulations, ML training, and surrogate optimization, is performed repeatedly to refine the solution in the vicinity of the predicted optimum. The proposed approach is applied to the optimization of a compression ignition engine with control parameters relating to fuel injection, in-cylinder flow, and thermodynamic conditions. It is demonstrated that by automatically selecting the best values of the hyperparameters, a 1.6% improvement in merit value is obtained, compared to an improvement of 1.0% with default hyperparameters. Overall, the framework introduced in this study reduces the need for technical expertise in training ML models for optimization, while also reducing the number of simulations needed for performing surrogate-based design optimization.

Download Full-text