Direct Fitness Research Articles

Surrogate-assisted multi-objective evolutionary feature selection of generation-based fixed evolution control for time series forecasting with LSTM networks

Surrogate-assisted multi-objective evolutionary algorithms are powerful techniques to solve computationally expensive multi-objective optimization problems. In this paper, we propose a direct fitness replacement method with generation-based fixed evolution control to implement a multi-objective evolutionary algorithm that uses a surrogate model for wrapper-type feature selection, where long short-term memory is established as the learning algorithm. The importance of the work and its benefits lie in the need to reduce the excessive computational time required by conventional wrapper-type feature selection methods based on multi-objective evolutionary algorithms and LSTM networks, maintaining or improving the predictive capacity of the models. We analyze the use of incremental learning to update the surrogate model, in comparison with the conventional non-incremental learning approach. We applied these methods in real-life time series forecasting of air quality, indoor temperature in a smart building and oil temperature in electricity transformers. Multi-step ahead predictions of the forecast models obtained with different meta-learners of the surrogate model were compared by using the Diebold–Mariano statistical test on a multi-criteria performance metric. The proposed method outperformed other approaches for feature selection including, among others, methods based on surrogate-assisted multi-objective evolutionary algorithms developed by the authors in previous research, other surrogate-assisted deterministic methods for feature selection and the conventional wrapper-type feature selection method based on LSTM, improving the prediction on test dataset by 23.98%, 34.61% and 13.77%, respectively.

Lifetime reproductive characteristics of gray wolves

Abstract Female and male cooperative breeders can use different strategies to maximize reproduction and fitness over their lifetimes. Answering questions about fitness in cooperative breeders requires long-term studies as well as complete data on group composition and size which can be exceedingly difficult to obtain. Using a long-term genetic data set of complete group pedigrees, I asked how lifetime reproductive characteristics of female and male gray wolves (Canis lupus) differed. I predicted that genetic relatedness to helpers would be higher for females than males due to philopatric behavior of female wolves, group size would be similar between the sexes, females would inherit breeding positions from within groups more often than males due to differences in dispersal strategies between the sexes, males would have more lifetime mates and produce more young than females because of polygamy, and females would breed for more years than males due to the likelihood that females would still breed (with a new partner) after a mate died or was expelled from the group. I documented complete lifetime breeding histories for 11 male and 18 female wolves in Idaho, United States, 2008 to 2018. Genetic relatedness to helpers, group size, number of mates, pups, and years breeding did not differ between the sexes. Females, however, inherited breeding positions within groups far more often than males. Individuals who secured breeding positions generally reproduced for 2 seasons and commonly had more than 1 partner during their lifetimes if they were able to maintain their breeding position longer. Direct fitness varied greatly within female and male breeding wolves.

Outbreeding reduces survival during metamorphosis in a headwater stream salamander.

Assessing direct fitness effects of individual genetic diversity is challenging due to the intensive and long-term data needed to quantify survival and reproduction in the wild. But resolving these effects is necessary to determine how inbreeding and outbreeding influence eco-evolutionary processes. We used 8 years of capture-recapture data and single nucleotide polymorphism genotypes for 1906 individuals to test for effects of individual heterozygosity on stage-specific survival probabilities in the salamander Gyrinophilus porphyriticus. The life cycle of G. porphyriticus includes an aquatic larval stage followed by metamorphosis into a semi-aquatic adult stage. In our study populations, the larval stage lasts 6-10 years, metamorphosis takes several months, and lifespan can reach 20 years. Previous studies showed that metamorphosis is a sensitive life stage, leading us to predict that fitness effects of individual heterozygosity would occur during metamorphosis. Consistent with this prediction, monthly probability of survival during metamorphosis declined with multi-locus heterozygosity (MLH), from 0.38 at the lowest MLH (0.10) to 0.06 at the highest MLH (0.38), a reduction of 84%. Body condition of larvae also declined significantly with increasing MLH. These relationships were consistent in the three study streams. With evidence of localised inbreeding within streams, these results suggest that outbreeding disrupts adaptations in pre-metamorphic and metamorphic individuals to environmental gradients along streams, adding to evidence that headwater streams are hotspots of microgeographic adaptation. Our results also underscore the importance of incorporating life history in analyses of the fitness effects of individual genetic diversity and suggest that metamorphosis and similar discrete life stage transitions may be critical periods of viability selection.

Observation of female-male mounting in the carrion crow

In the biological sciences, sexual behaviours in non-human animals are traditionally investigated in the context of reproduction and direct fitness benefits. While the evolutionary functions of non-conceptive sexual behaviours (‘socio-sexual behaviours’) remain less well explored, these interactions and displays have been suggested to be important for shaping and maintaining social relationships. Here, we report an observation of a captive female carrion crow, Corvus corone corone, mounting her co-housed male partner. We highlight the importance of more systematic research, reporting, and discussions of rarely observed behaviours in social evolution research, including considerations for behaviours that transcend binary or heteronormative frameworks, for a more comprehensive understanding of non-conceptive socio-sexual behaviours.

Direct fitting improves the accuracy of the horse radish peroxidase competition assay for peroxidase activity

The peroxiredoxins are an important antioxidant protein family and their ability to neutralise oxidants is regularly investigated using horse radish peroxidase in a competition assay system. In this method, the rate constant of a peroxiredoxin is calculated from the fractional inhibition of horse radish peroxidase activity caused by competition with the peroxiredoxin for an oxidant substrate. We developed a model capable of simulating this assay and, using this model, demonstrate that the fractional inhibition calculation significantly and systematically mis-estimates the rate constant under fairly common conditions. We go on to develop a method for fitting simulated assay time-courses to experimental data directly, which significantly outperforms the fractional inhibition method yielding more accurate results. Based on our findings, we recommend using the direct fitting approach to determine peroxidase rate constants from horseradish peroxidase experiments.

Learning-based Free-Water Correction using Single-shell Diffusion MRI.

Diffusion magnetic resonance imaging (dMRI) offers the ability to assess subvoxel brain microstructure through the extraction of biomarkers like fractional anisotropy, as well as to unveil brain connectivity by reconstructing white matter fiber trajectories. However, accurate analysis becomes challenging at the interface between cerebrospinal fluid and white matter, where the MRI signal originates from both the cerebrospinal fluid and the white matter partial volume. The presence of free water partial volume effects introduces a substantial bias in estimating diffusion properties, thereby limiting the clinical utility of DWI. Moreover, current mathematical models often lack applicability to single-shell acquisitions commonly encountered in clinical settings. Without appropriate regularization, direct model fitting becomes impractical. We propose a novel voxel-based deep learning method for mapping and correcting free-water partial volume contamination in DWI to address these limitations. This approach leverages data-driven techniques to reliably infer plausible free-water volumes across different diffusion MRI acquisition schemes, including single-shell acquisitions. Our evaluation demonstrates that the introduced methodology consistently produces more consistent and plausible results than previous approaches. By effectively mitigating the impact of free water partial volume effects, our approach enhances the accuracy and reliability of DWI analysis for single-shell dMRI, thereby expanding its applications in assessing brain microstructure and connectivity.

Research on etalon‐free CO2 measurement based on direct absorption signal fitting

AbstractThe emission of carbon dioxide (CO2) by thermal power plants is the main factor leading to the greenhouse effect. The measurement of CO2 is of great significance to carbon emission data statistics. In this paper, a measurement system based on tunable diode laser absorption spectroscopy (TDLAS) was built to measure the spectrum of CO2 near 1579 nm. To simplify the experimental system, a direct absorption signal fitting method is proposed to replace etalons for time‐frequency conversion. The method was applied to the measurement of four different concentrations of CO2 in the range of 7%–16%, and the measurement results were compared with the multipeak method, another etalon‐free method. The results show that the mean relative error of the measurement results is 4.23%, which is 1.76% lower than the measurement results of the multipeak method. The experimental research provides a feasible idea to simplify the CO2 measurement system based on TDLAS.

Numerical investigations on constitutive model parameters of HRB400 and HTRB600 steel bars based on tensile and fatigue tests

Abstract In recent years, HRB400 and HTRB600 steel bars have become the mainstream standard reinforcing steel used in concrete structures in China. However, significant controversy still exists regarding the selection of material constitutive models and the determination of model parameters for buckling, fatigue, hysteresis, and other material characteristics. In this article, an automated process of multi-parameter calculation of the constitutive model for reinforcing steel – simulation accuracy evaluation of the constitutive model – selection of the constitutive model of reinforcing steel is established based on the hybrid programming method using MATLAB and OpenSees software. First, tensile and low-cycle fatigue tests were carried out on HRB400 and HTRB600 steel bars. Second, based on the constitutive model in OpenSees software and the skeleton curve and characteristics such as yielding, fatigue, and hysteresis, the constitutive model parameters of HRB400 and HTRB600 steel bars are determined using indirect and direct fitting methods. Finally, the five similarity parameters of the simulated normalized cumulative hysteretic energy dissipation coefficient are compared with the test results. The results indicate that the simulation accuracy of the Reinforcing Steel model exceeds 72%, which is higher than other four models, making it the best choice for reinforcing steel in numerical simulation.

Direct Discrepancy Dynamic Fit Index Cutoffs for Arbitrary Covariance Structure Models

Despite the popularity of traditional fit index cutoffs like RMSEA ≤ .06 and CFI ≥ .95, several studies have noted issues with overgeneralizing traditional cutoffs. Computational methods have been proposed to avoid overgeneralization by deriving cutoffs specifically tailored to the characteristics of the model being evaluated. Simulations show favorable performance of these methods; however, these methods support a narrow set of scenarios (e.g., certain models or response scales) and the interpretation of cutoffs is not always standardized, which affects empirical researchers’ ability to confidently and broadly adopt these methods to evaluate model fit. In this paper, we propose an extension to one recently developed computational method—dynamic fit index cutoffs—that (a) permits application to any covariance structure model (e.g., CFA, mediation, bifactor), (b) standardizes interpretation of cutoffs across any covariance structure model, and (c) supports normal, nonnormal, categorical, and missing data. Software is provided to facilitate implementation of the method.

Inverse problems in resonant ultrasound spectroscopy based on spectra perturbation

In RUS, the fit between the experimental resonant spectrum and the result of the inversion procedure is usually worse than the inaccuracy of the experimental measurement of the resonant frequencies. Thus, it can be problematic to extract any valuable information from slight perturbation of experimental spectra, where the experimental frequency shift falls well below the accuracy of this fit. However, the inverse procedure can be reformulated and instead of fitting the resonant frequencies itself, it can be based on the direct fitting of the frequency perturbations. In this contribution, we show two examples of the use of this approach (i) quantitative RUS evaluation of the elasticity of micrometric and submicrometric surface layers from frequency differences measured before and after layer deposition [1] and (ii) in situ characterization of the thermal degradation of the bonding layer in a bimetallic system [2]. [1] Thomasová M. et al. Young's moduli of sputter-deposited NiTi films determined by resonant ultrasound spectroscopy: Austenite, R-phase, and martensite (2015) Scripta Materialia, 101, pp. 24–27. [2] Janovská M. et al Characterization of bonding quality of a cold-sprayed deposit by laser resonant ultrasound spectroscopy (2020) Ultrasonics, 106, art. no. 106140.

A validation study of the Occupational Depression Inventory in Poland and Ukraine

This study examined the psychometric and structural properties of the Polish and Ukrainian versions of the Occupational Depression Inventory (ODI). We relied on two samples of Polish employees (NSample1 = 526, 47% female; NSample2 = 164, 64% female) and one sample of Ukrainian employees (NSample3 = 372, 73% female). In all samples, the ODI exhibited essential unidimensionality and high total-score reliability (e.g., McDonald’s omegas > 0.90). The homogeneity of the scale was strong (e.g., 0.59 ≤ scale-level Hs ≤ 0.68). The ODI’s total scores thus accurately ranked individuals on a latent occupational depression continuum. We found evidence of complete measurement invariance across our samples, a prerequisite for between-group comparisons involving observed scores. Looking into the criterion validity of the ODI, we found occupational depression to correlate, in the expected direction, with resilience and job-person fit in six areas of working life—workload, control, rewards, community, fairness, and values. The prevalence of occupational depression was estimated at 5% in Sample 1, 18% in Sample 2, and 3% in Sample 3. Our findings support the use of the ODI’s Polish and Ukrainian versions. This study adds to a growing corpus of research suggesting that the ODI is a robust instrument.

Gametic selection favours polyandry and selfing.

Competition among pollen or sperm (gametic selection) can cause evolution. Mating systems shape the intensity of gametic selection by determining the competitors involved, which can in turn cause the mating system itself to evolve. We model the bidirectional relationship between gametic selection and mating systems, focusing on variation in female mating frequency (monandry-polyandry) and self-fertilisation (selfing-outcrossing). First, we find that monandry and selfing both reduce the efficiency of gametic selection in removing deleterious alleles. This means that selfing can increase mutation load, in contrast to cases without gametic selection where selfing purges deleterious mutations and decreases mutation load. Second, we explore how mating systems evolve via their effect on gametic selection. By manipulating gametic selection, polyandry can evolve to increase the fitness of the offspring produced. However, this indirect advantage of post-copulatory sexual selection is weak and is likely to be overwhelmed by any direct fitness effects of mating systems. Nevertheless, gametic selection can be potentially decisive for selfing evolution because it significantly reduces inbreeding depression, which favours selfing. Thus, the presence of gametic selection could be a key factor driving selfing evolution.

Theoretical Investigation of Low Global Warming Potential Blends Replacing R404A: The Simple Refrigeration Cycle and Its Modifications

Abstract R404A refrigerant is one of the most widely used hydrofluorocarbons in the commercial refrigeration industry for low and medium temperatures. However, this refrigerant contributes negatively to the environment due to its high global warming potential (GWP = 3943), and for several years, it has been labeled as one of the refrigerants that should be phased out. The present study theoretically evaluates low-GWP refrigerants as a replacement alternative to R404A, including R407H, R442A, R449A, R454A, R454C, R455A, R459B, and R465A. The analysis focuses on the comparison between relative differences in the volumetric flow in the compressor suction and the coefficient of performance for four configurations of the vapor compression cycle, such as the basic cycle, the cycle with an internal heat exchanger, the cycle with direct injection, and the cycle with a sub-cooler. According to the proposed operating conditions of evaporation temperature (−10 °C and −40 °C) and condensation temperature (40 °C and 55 °C), R454A could be the best long-term replacement option for its low GWP, energy performance, and direct fit for any configuration. R459B could also be considered a viable option, but with certain design modifications. On the contrary, the refrigerants R465A, R455A, and R454C would be discarded because they present a greater non-adaptation to the compressor in each configuration analyzed.

Comparison of methods for intravoxel incoherent motion parameter estimation in the brain from flow-compensated and non-flow-compensated diffusion-encoded data.

Joint analysis of flow-compensated (FC) and non-flow-compensated (NC) diffusion MRI (dMRI) data has been suggested for increased robustness of intravoxel incoherent motion (IVIM) parameter estimation. For this purpose, a set of methods commonly used or previously found useful for IVIM analysis of dMRI data obtained with conventional diffusion encoding were evaluated in healthy human brain. Five methods for joint IVIM analysis of FC and NC dMRI data were compared: (1) direct non-linear least squares fitting, (2) a segmented fitting algorithm with estimation of the diffusion coefficient from higher b-values of NC data, (3) a Bayesian algorithm with uniform prior distributions, (4) a Bayesian algorithm with spatial prior distributions, and (5) a deep learning-based algorithm. Methods were evaluated on brain dMRI data from healthy subjects and simulated data at multiple noise levels. Bipolar diffusion encoding gradients were used with b-values 0-200 s/mm2 and corresponding flow weighting factors 0-2.35 s/mm for NC data and by design 0 for FC data. Data were acquired twice for repeatability analysis. Measurement repeatability as well as estimation bias and variability were at similar levels or better with the Bayesian algorithm with spatial prior distributions and the deep learning-based algorithm for IVIM parameters and , and for the Bayesian algorithm only for , relative to the other methods. A Bayesian algorithm with spatial prior distributions is preferable for joint IVIM analysis of FC and NC dMRI data in the healthy human brain, but deep learning-based algorithms appear promising.

Parameter-Free and Electron Counting Satisfied Material Representation for Machine Learning Potential Energy and Force Fields.

We proposed a parameter-free volume element representation that satisfies the electron counting model and obtains accurate machine learning potential energy and direct force fitting of randomly perturbed hexagonal BN. Our method preserves permutational, translational, and rotational invariance and can be extended to three-dimensional systems, verified by a system of bulk Si. As a result, we obtained 0.57 meV/atom potential energy root mean squared error (RMSE) and 59 meV/Å force RMSE for perturbed bulk BN systems and 0.43 meV/atom potential energy RMSE and 36 meV/Å force RMSE for perturbed Si systems. In addition, an unbiased perturbation-based data set construction scheme is introduced and a continuous population distribution is obtained with a training data set of 4500, which is about 1 order of magnitude smaller than standard methods based on first-principles molecular dynamics simulations and saves a large amount of computing resources. General validity of our model is verified by structure optimization, molecular dynamics simulations, and extrapolations.

Little evidence for color- or size-based mating preferences by male strawberry poison frogs (Oophaga pumilio)

While there are many studies documenting female mating preferences across taxa, male mate choice remains relatively understudied. Male mate choice often develops when there is variation in female quality and thus the fitness benefits of mating with particular females. Specifically, males tend to prefer females with traits that confer direct fitness benefits such as large body size, which may be linked with high fecundity. Prior work has shown that females of the strawberry poison frog, Oophaga pumilio, prefer males bearing certain coloration (most often the female’s own color), and that this preference can be learned through maternal imprinting. Females have been shown to prefer larger males as well. Here we test whether similar mate preferences for color and size exist in males of this species using two-way choice tests on captive bred male O. pumilio. In each test focal males were placed in an arena with two stimulus females: either both of the same size but differing in color, or both of the same color but differing in size. We found only weak evidence for behavioral biases toward particular colors and no evidence for biases toward larger females, suggesting that males of O. pumilio do not predictably choose mates based on these female traits. Despite several aspects of their natural history that suggest males have reasons to be choosy, our findings suggest that the cost of mate rejection may outweigh any fitness benefits derived from being selective of mates. Studies of additional populations, ideally conducted on wild individuals, are needed to better understand the range of conditions under which males may exhibit mate choice and the types of traits on which they base these choices.Significance statementTo fully understand the fitness landscapes and evolutionary trajectories that result from sexual selection, we need to understand when and how the mate preferences of the two sexes act and interact. While female mate choice has been widely studied, male mate choice remains poorly understood. To help bridge this gap, we studied male mate preferences in the strawberry poison frog Oophaga pumilio, a small brightly colored frog for which female preferences for male color and size have been well-documented. We found no evidence that male O. pumilio exhibit mate preferences based on female size and little evidence for male mate preferences based on female color. This is surprising given that larger females are often more fecund, male O. pumilio are known to exhibit color-based behavioral biases in the context of male-male competition, and both sexes provide parental care.

Direct Fit for SVI Implied Volatilities

Direct Fit for SVI Implied Volatilities

Saltire: a model to measure dynamical masses for high-contrast binaries and exoplanets with high-resolution spectroscopy

ABSTRACT High-resolution cross-correlation methods are widely used to discover and to characterise atomic and molecular species in exoplanet atmospheres. The characteristic cross-correlation signal is typically represented as a function of the velocity of the system, and the semi-amplitude of the planet’s orbit. We present Saltire, a fast and simple model that accurately reproduces the shape of such cross-correlation signals, allowing a direct fit to the data by using a minimum set of parameters. We show how to use this model on the detection of atmospheric carbon monoxide in archival data of the hot Jupiter $\rm \tau$ Boötis b, and how Saltire can be used to estimate the semi-amplitude and rest velocity of high brightness ratio binaries. By including the shape of the signal, we demonstrate that our model allows to robustly derive the signal position up to 10 times more accurate, compared to conventional methods. Furthermore, we discuss the impact of correlated noise and demonstrate that Saltire is a robust tool for estimating systematic uncertainties on the signal position. Saltire opens a new door to analyse high signal-to-noise data to accurately study atmospheric dynamics and to measure precise dynamical masses for exoplanets and faint stellar companions. We show that the phase-resolved shape of the atmospheric cross-correlation signal can accurately be reproduced, allowing studies of phase-dependent signal changes and to disentangle them from noise and data aliases.

The ground electronic state of CS: A global multi-isotopologue direct potential fit analysis

A collection of 6759 pure-rotational and vibrational–rotational spectroscopic line positions available for seven isotopologues of the ground X1Σ+ electronic state of carbon monosulfide has been employed in a weighted nonlinear least-squares direct fit to the potential energy function of 12C32S and supplementary functions describing breakdown of the Born-Oppenheimer approximation. All radial functions are represented by compact analytical models having proper theoretical asymptotic behavior. The spectroscopic line positions are reproduced on average to within the associated experimental uncertainties by the quantum–mechanical eigenvalues of the derived Hamiltonians for individual isotopologues. The potential energy function is constrained to obey the theoretical radial behavior both at short-range and at long-range. Accurate quantum–mechanical vibrational term values and rotational and centrifugal distortion constants are provided for all stable isotopologues of CS included in the least-squares fits.

Large fitness benefits of social nesting in a small carpenter bee

Abstract Facultatively social insects are an optimal model group for the study of the emergence of cooperation between individuals. Factors influencing the fitness benefits of social nesting are still debated. Nonreproductive subordinates can benefit from indirect fitness benefits due to increasing reproductive success of related individuals or direct fitness benefits due to direct future reproduction. Here, we studied the costs and benefits of social nesting in the small carpenter bee Ceratina albosticta. From demographic data and within-nest relatedness, we obtained key parameters for assessing the fitness of solitary females, social primaries, and social secondaries. C. albosticta were found to usually mate with one male and multiple mating is not common although exists at low frequencies. Social nests usually contain two females, which were found to be related (full sisters), but also some females were unrelated to each other. Patterns of parentage from microsatellite loci revealed that only one female reproduces in social nests. Our results show that relatedness, per capita brood productivity, and offspring survival strongly increase the fitness benefits of social nesting strategies. Social secondaries, when related to the social primaries, have higher inclusive fitness than solitary females, but unrelated social nesting females had no indirect fitness and much reduced inclusive fitness compared to solitary females. Interestingly, average fitness benefits of the social secondary were higher than solitary females. This study provides important empirical data on the costs and benefits of sociality in a facultatively social bee and sets the stage for future comparative studies.