Selection Gradients Research Articles

Integrating Health Data-Driven Machine Learning Algorithms to Evaluate Risk Factors of Early Stage Hypertension at Different Levels of HDL and LDL Cholesterol.

Purpose: Cardiovascular disease (CVD) is a major worldwide health burden. As the risk factors of CVD, hypertension, and hyperlipidemia are most mentioned. Early stage hypertension in the population with dyslipidemia is an important public health hazard. This study was the application of data-driven machine learning (ML), demonstrating complex relationships between risk factors and outcomes and promising predictive performance with vast amounts of medical data, aimed to investigate the association between dyslipidemia and the incidence of early stage hypertension in a large cohort with normal blood pressure at baseline. Methods: This study analyzed annual health screening data for 71,108 people from 2005 to 2017, including data for 27 risk-related indicators, sourced from the MJ Group, a major health screening center in Taiwan. We used five machine learning (ML) methods—stochastic gradient boosting (SGB), multivariate adaptive regression splines (MARS), least absolute shrinkage and selection operator regression (Lasso), ridge regression (Ridge), and gradient boosting with categorical features support (CatBoost)—to develop a multi-stage ML algorithm-based prediction scheme and then evaluate important risk factors at the early stage of hypertension, especially for groups with high-density lipoprotein cholesterol (HDL-C) and low-density lipoprotein cholesterol (LDL-C) levels within or out of the reference range. Results: Age, body mass index, waist circumference, waist-to-hip ratio, fasting plasma glucose, and C-reactive protein (CRP) were associated with hypertension. The hemoglobin level was also a positive contributor to blood pressure elevation and it appeared among the top three important risk factors in all LDL-C/HDL-C groups; therefore, these variables may be important in affecting blood pressure in the early stage of hypertension. A residual contribution to blood pressure elevation was found in groups with increased LDL-C. This suggests that LDL-C levels are associated with CPR levels, and that the LDL-C level may be an important factor for predicting the development of hypertension. Conclusion: The five prediction models provided similar classifications of risk factors. The results of this study show that an increase in LDL-C is more important than the start of a drop in HDL-C in health screening of sub-healthy adults. The findings of this study should be of value to health awareness raising about hypertension and further discussion and follow-up research.

Modelling of aerobic granular sludge reactors: the importance of hydrodynamic regimes, selective sludge removal and gradients.

Hydraulic selection is a key feature of aerobic granular sludge (AGS) systems but existing aerobic granular sludge (AGS) models neglect those mechanisms: gradients over reactor height (Hreactor), selective removal of slow settling sludge, etc. This study aimed at evaluating to what extent integration of those additional processes into AGS models is needed, i.e., at demonstrating that model predictions (biomass inventory, microbial activities and effluent quality) are affected by such additional model complexity. We therefore developed a new AGS model that includes key features of full-scale AGS systems: fill-draw operation, selective sludge removal, distinct settling models for flocs/granules. We then compared predictions of our model to those of a fully mixed AGS model. Our results demonstrate that hydraulic selection can be predicted with an assembly of four continuous stirred tank reactors in series together with a correction code for plug-flow. Concentration gradients over the reactor height during settling/plug-flow feeding strongly impact the predictions of aerobic granular sludge models in terms of microbial selection, microbial activities and ultimately effluent quality. Hydraulic selection is a key to predict selection of storing microorganisms (phosphorus-accumulating organisms (PAO) and glycogen-accumulating organisms (GAO)) and in turn effluent quality in terms of total phosphorus, and for predicting effluent solid concentration and dynamic during plug-flow feeding.

Comprehensive Evaluation and Comparison of Machine Learning Methods in QSAR Modeling of Antioxidant Tripeptides.

Due to their multiple beneficial effects, antioxidant peptides have attracted increasing interest. Currently, the screening and identification of bioactive peptides, including antioxidative peptides based on wet-chemistry methods are time-consuming and highly rely on many advanced instruments and trained personnel. Quantitative structure–activity relationship (QSAR) analysis as an in silico method can be more efficient and cost-effective. However, model performance of QSAR studies on antioxidant peptides was still poor due to limited attempts in model development approaches. The objective of this study was to compare popular machine learning methods for antioxidant activity modeling and screening of tripeptides and identify the critical amino acid features that determine the antioxidant activity. 533 numerical indices of amino acids were adopted to characterize 130 tripeptides with known antioxidant activity from the published literature, and then 7 feature selection strategies plus pairwise correlation were used to screen the most important indices for antioxidant activity and model building. 14 machine learning methods were used to build models based on the feature selection strategies, respectively. Among the 98 models, non-linear regression methods tended to perform better, and the best model with an R2Test of 0.847 and RMSETest of 0.627 for tripeptide antioxidants was obtained by combining random forest for feature selection and tree-based extreme gradient boost regression for model development. Based on the predicted antioxidant values of 7870 unknown tripeptides, potentially high antioxidant activity tripeptides all have a tyrosine, tryptophan, or cysteine residue at the C-terminal position. Furthermore, the predicted antioxidant activity of six synthesized tripeptides was confirmed through experimental determination, and for the first time, the cysteine or tyrosine residue at the C-terminal was found to be critical to the antioxidant activity based on both QSAR models and experimental observations.

CT-Based Radiomics Nomogram Improves Risk Stratification and Prediction of Early Recurrence in Hepatocellular Carcinoma After Partial Hepatectomy.

ObjectivesTo develop and validate an intuitive computed tomography (CT)-based radiomics nomogram for the prediction and risk stratification of early recurrence (ER) in hepatocellular carcinoma (HCC) patients after partial hepatectomy.MethodsA total of 132 HCC patients treated with partial hepatectomy were retrospectively enrolled and assigned to training and test sets. Least absolute shrinkage and selection operator and gradient boosting decision tree were used to extract quantitative radiomics features from preoperative contrast-enhanced CT images of the HCC patients. The radiomics features with predictive value for ER were used, either alone or in combination with other predictive features, to construct predictive models. The best performing model was then selected to develop an intuitive, simple-to-use nomogram, and its performance in the prediction and risk stratification of ER was evaluated using the area under the receiver operating characteristic curve (AUC), calibration curve, and decision curve analysis (DCA).ResultsThe radiomics model based on the radiomics score (Rad-score) achieved AUCs of 0.870 and 0.890 in the training and test sets, respectively. Among the six predictive models, the combined model based on the Rad-score, Edmondson grade, and tumor size had the highest AUCs of 0.907 in the training set and 0.948 in the test set and was used to develop an intuitive nomogram. Notably, the calibration curve and DCA for the nomogram showed good calibration and clinical application. Moreover, the risk of ER was significantly different between the high- and low-risk groups stratified by the nomogram (p <0.001).ConclusionsThe CT-based radiomics nomogram developed in this study exhibits outstanding performance for ER prediction and risk stratification. As such, this intuitive nomogram holds promise as a more effective and user-friendly tool in predicting ER for HCC patients after partial hepatectomy.

Evolution of class-structured populations in periodic environments.

What is the influence of periodic environmental fluctuations on life‐history evolution? We present a general theoretical framework to understand and predict the long‐term evolution of life‐history traits under a broad range of ecological scenarios. Specifically, we investigate how periodic fluctuations affect selection when the population is also structured in distinct classes. This analysis yields time‐varying selection gradients that clarify the influence of the fluctuations of the environment on the competitive ability of a specific life‐history mutation. We use this framework to analyse the evolution of key life‐history traits of pathogens. We examine three different epidemiological scenarios and we show how periodic fluctuations of the environment can affect the evolution of virulence and transmission as well as the preference for different hosts. These examples yield new and testable predictions on pathogen evolution, and illustrate how our approach can provide a better understanding of the evolutionary consequences of time‐varying environmental fluctuations in a broad range of scenarios.

Stratification in parents’ selection of developmentally appropriate books for children: register-based evidence from Danish public libraries

ABSTRACT This paper studies socioeconomic gradients in parents' selection of developmentally appropriate children's books from public libraries. I draw on developmental gradients research to hypothesize that families with high socioeconomic status (SES) select a higher share of books that match children's developmental stage in order to best improve children's learning environments. Based on Danish registry data on all books borrowed from public libraries in 2020, I find that highly educated families are more likely to use libraries and borrow more books, but they do not select a larger share of developmentally appropriate books; in fact, they select a slightly lower share. In contrast, high-income families borrow both a little more books and a little higher share of developmentally appropriate books, than low-income families. The supplementary analyses show that results are robust across families with children of different ages and to account for nonrandom selection into the sample of library users, socioeconomic differences in children's reading skills, and the impact of library lockdowns due to Covid-19. I conclude that stratification in library book selection is more prominent concerning the voraciousness with which highly educated parents provide reading inputs (more books) than how discriminating they are in terms of selecting developmentally appropriate books.

Eco-evolutionary dynamics of intergenomic epistatic QTLs under slight and hard multilevel selections in community genetics

Eco-evolutionary community genetics refers to the study of population genetics at multispecies levels since a single species evolves genetically through interactions with other species. The aim of this paper is dynamical modeling of interspecific quantitative trait loci (QTL) under slight and hard multilevel selection to investigate eco-evolutionary genetic relationships among QTL of multispecies. It was found that the simplex explained the intergenomic epistatic QTL (IEQ) dynamics in a discrete-time model according to Wright’s manifold. Besides, the generalized gradient system on the simplex demonstrated the IEQ dynamics in a continuous-time model. The outcomes of slight and hard multilevel selection on the expressions of IEQ in species were explained by the gradient operator of the mean multilevel fitness. IEQ analysis was integrated into community genetics according to Lande’s assumptions, so for slight multilevel selection, the geometric average of the mean multilevel fitness specified Wright’s idea of adaptive topography. The dynamical system for phenotypic eco-evolution under a hard multilevel selection gradient was studied and the relative mean multilevel fitness of two species was estimated.

VPFL: A verifiable privacy-preserving federated learning scheme for edge computing systems

Federated learning for edge computing is a promising solution in the data booming era, which leverages the computation ability of each edge device to train local models and only shares the model gradients to the central server. However, the frequently transmitted local gradients could also leak the participants’ private data. To protect the privacy of local training data, lots of cryptographic-based Privacy-Preserving Federated Learning (PPFL) schemes have been proposed. However, due to the constrained resource nature of mobile devices and complex cryptographic operations, traditional PPFL schemes fail to provide efficient data confidentiality and lightweight integrity verification simultaneously. To tackle this problem, we propose a Verifiable Privacy-preserving Federated Learning scheme (VPFL) for edge computing systems to prevent local gradients from leaking over the transmission stage. Firstly, we combine the Distributed Selective Stochastic Gradient Descent (DSSGD) method with Paillier homomorphic cryptosystem to achieve the distributed encryption functionality, so as to reduce the computation cost of the complex cryptosystem. Secondly, we further present an online/offline signature method to realize the lightweight gradients integrity verification, where the offline part can be securely outsourced to the edge server. Comprehensive security analysis demonstrates the proposed VPFL can achieve data confidentiality, authentication, and integrity. At last, we evaluate both communication overhead and computation cost of the proposed VPFL scheme, the experimental results have shown VPFL has low computation costs and communication overheads while maintaining high training accuracy.

Analytical results for directional and quadratic selection gradients for log-linear models of fitness functions.

Log‐linear models are widely used for assessing determinants of fitness in empirical studies, for example, in determining how reproductive output depends on trait values or environmental conditions. Similarly, theoretical works of fitness and natural selection employ log‐linear models, often with a negative quadratic term, generating Gaussian fitness functions. However, in the specific application of regression‐based analysis of natural selection, such models are rarely employed. Rather, OLS regression is the predominant means of assessing the form of natural selection. OLS regressions allow specific evolutionary quantitative parameters, selection gradients, to be estimated, and benefit from the fact that the associated statistical models are easily applied. We examine whether selection gradients can be directly expressed in terms of the coefficients of models using exponential fitness functions with linear or quadratic arguments. Such models can be easily fitted with generalized linear models (GLMs). The expressions we obtain coincide with those for Gaussian functions, but relax the major constraint that the (log) fitness function is concave (downwardly curved). Additionally these results lead to univariate and multivariate analyses of both linear and quadratic selection that potentially incorporate pragmatic and interpretable models of fitness functions, where the parameters can be related analytically to selection gradients, and that can be operationalized using widely available statistical tools.

Abstract B002: Modeling evolutionary therapy for GI malignancies

Abstract Despite development of personalized treatments GI cancer like pancreatic ductal adenocarcinoma (PDAC) and colorectal cancer (CRC) remain dismal outcomes. While some patients are potentially curable with surgery and chemotherapy, most unresectable cases are invariably lethal. To advance patient care, we rely on clinical trials to test new therapeutic approaches. Here, we assume that the highest dose of a drug that does not cause severe toxicities is optimal for cancer treatment. However, this treatment strategy, in many cases, may be more harmful for the patient. Adopting principles from the field of evolutionary biology and pest management could offer more effective strategies that can be used for patients with metastatic PDAC and CRC. Resistance mechanisms are assumed to come with an energy cost, making the organism/cell less fit than others in the population. When high doses of treatment are applied, it results in a significant selection gradient by eliminating sensitive populations, freeing resources and facilitating resistant population growth. Such evolutionary dynamics and interactions within the tumor microenvironment can be modeled and exploited, in the form of adaptive strategies, for treatment of CRC and PDAC. It is our central hypothesis that integration of evolutionary dynamics and into PDAC and CRC treatment will delay emergence of resistance, enhance quality of life, and improve clinical outcomes. Our preliminary observations in co-cultured drug sensitive and resistant cell containing spheroids show that a higher proportion of drug sensitive cells inhibits the proliferation of drug resistant cells over time, in the presence of chemotherapeutic agents 5-FU and Oxaliplatin. We also demonstrate a role for the MLL family of histone methyltransferases in the epigenetic regulation of genes involved in chemoresistance. RNA-seq data has shown differential gene expression in metabolic genes regulated by histone methyltransferase KMT2A, KM2TB and KMT2D knockout (KO) cells including Gstt1, CYP26B1 and PYGL. In addition to this, KMT2A, KM2TB and KMT2D KOs show differential sensitivity to chemotherapeutic agents Gemcitabine, SN-38, 5-FU and Oxaliplatin. Further, analysis of KMT2A, KM2TB and KMT2D KO cells will provide mechanistic insight of the epigenetic regulation of metabolic patterns in drug resistant cells. This strategy would provide an approach for reducing exposure to toxic drugs, improving patient quality of life and renewing hope for prolonged control of metastatic PDAC and CRC. This proposal represents a bedside-to-bench-to-bedside project; identifying a shortcoming in how we currently treat PDAC and CRC patients, developing solutions in the lab and ultimately translating such strategies to benefit patients. Citation Format: Glancis Luzeena Raja Arul, Ezequiel Tolosa, Ryan M. Carr, Martin E. Fernandez-Zapico. Modeling evolutionary therapy for GI malignancies [abstract]. In: Proceedings of the AACR Special Conference on the Evolutionary Dynamics in Carcinogenesis and Response to Therapy; 2022 Mar 14-17. Philadelphia (PA): AACR; Cancer Res 2022;82(10 Suppl):Abstract nr B002.

Reduced olfactory acuity in recently flightless insects suggests rapid regressive evolution

BackgroundInsects have exceptionally fast smelling capabilities, and some can track the temporal structure of odour plumes at rates above 100 Hz. It has been hypothesized that this fast smelling capability is an adaptation for flying. We test this hypothesis by comparing the olfactory acuity of sympatric flighted versus flightless lineages within a wing-polymorphic stonefly species.ResultsOur analyses of olfactory receptor neuron responses reveal that recently-evolved flightless lineages have reduced olfactory acuity. By comparing flighted versus flightless ecotypes with similar genetic backgrounds, we eliminate other confounding factors that might have affected the evolution of their olfactory reception mechanisms. Our detection of different patterns of reduced olfactory response strength and speed in independently wing-reduced lineages suggests parallel evolution of reduced olfactory acuity.ConclusionsThese reductions in olfactory acuity echo the rapid reduction of wings themselves, and represent an olfactory parallel to the convergent phenotypic shifts seen under selective gradients in other sensory systems (e.g. parallel loss of vision in cave fauna). Our study provides evidence for the hypothesis that flight poses a selective pressure on the speed and strength of olfactory receptor neuron responses and emphasizes the energetic costs of rapid olfaction.

Divergent selection on behavioural and chemical traits between reproductively isolated populations of Drosophila melanogaster.

Speciation is driven by traits that can act to prevent mating between nascent lineages, including male courtship and female preference for male traits. Mating barriers involving these traits evolve quickly because there is strong selection on males and females to maximize reproductive success, and the tight co‐evolution of mating interactions can lead to rapid diversification of sexual behaviour. Populations of Drosophila melanogaster show strong asymmetrical reproductive isolation that is correlated with geographic origin. Using strains that capture natural variation in mating traits, we ask two key questions: which specific male traits are females selecting, and are these traits under divergent sexual selection? These questions have proven extremely challenging to answer, because even in closely related lineages males often differ in multiple traits related to mating behaviour. We address these questions by estimating selection gradients for male courtship and cuticular hydrocarbons for two different female genotypes. We identify specific behaviours and particular cuticular hydrocarbons that are under divergent sexual selection and could potentially contribute to premating reproductive isolation. Additionally, we report that a subset of these traits are plastic; males adjust these traits based on the identity of the female genotype they interact with. These results suggest that even when male courtship is not fixed between lineages, ongoing selection can act on traits that are important for reproductive isolation.

A novel feature selection with stochastic gradient descent logistic regression for multilabeled stress prediction in working employees

AbstractIn the recent times, stress prediction becomes a hot research area and several research works have been developed to address it. The advent of machine learning (ML) models assists the stress prediction process to understand the patterns effectively and offer effective perceptions about possible future intervention. In this view, this article presents a multi‐labeled stress prediction in working employee using extremely randomized tree (ET) based feature selection (FS) and stochastic gradient descent (SGD) with logistic regression (LR), called ETSGD‐LR model. First, the ET based FS technique can be used to compute impurity‐based feature importance, which in turn can be used to discard irrelevant features. In addition, the SGD‐LR model is used to classify the feature reduced subset into different class labels. For experimental validation, we have collected our own stress prediction dataset with 1197 records of employees collected from schools, banks, universities, and so forth from different institutions. Among them 1197 records are filtered with various diseases and work pressure. A detailed set of simulations were carried out in Python Programming tool, and the results are analyzed in terms of sensitivity, specificity, accuracy, precision, F‐score, and kappa. The obtained simulation outcome ensured the superior performance of the ETSGD‐LR model over the compared methods with the maximum sensitivity, specificity, and accuracy of 0.980, 0.900, and 0.972, respectively. The experimental results shown that the inclusion of FS process helps to improve the classification performance.

Phenotypic selection on floral traits in the arctic plant Parrya nudicaulis (Brassicaceae).

The evolution of floral traits is often attributed to pollinator‐mediated selection; however, the importance of pollinators as selective agents in arctic environments is poorly resolved. In arctic and subarctic regions that are thought to be pollen limited, selection is expected to either favor floral traits that increase pollinator attraction or promote reproductive assurance through selfing. We quantified phenotypic selection on floral traits in two arctic and two subarctic populations of the self‐compatible, but largely pollinator‐dependent, Parrya nudicaulis. Additionally, we measured selection in plants in both open pollination and pollen augmentation treatments to estimate selection imposed by pollinators in one population. Seed production was found to be limited by pollen availability and strong directional selection on flower number was observed. We did not detect consistently greater magnitudes of selection on floral traits in the arctic relative to the subarctic populations. Directional selection for more pigmented flowers in one arctic population was observed, however. In some populations, selection on flower color was found to interact with other traits. We did not detect consistently stronger selection gradients across all traits for plants exposed to pollinator selection relative to those in the pollen augmentation treatment; however, directional selection tended to be higher for some floral traits in open‐pollinated plants.

A Synthesis of Game Theory and Quantitative Genetic Models of Social Evolution.

Two popular approaches for modeling social evolution, evolutionary game theory and quantitative genetics, ask complementary questions but are rarely integrated. Game theory focuses on evolutionary outcomes, with models solving for evolutionarily stable equilibria, whereas quantitative genetics provides insight into evolutionary processes, with models predicting short-term responses to selection. Here we draw parallels between evolutionary game theory and interacting phenotypes theory, which is a quantitative genetic framework for understanding social evolution. First, we show how any evolutionary game may be translated into two quantitative genetic selection gradients, nonsocial and social selection, which may be used to predict evolutionary change from a single round of the game. We show that synergistic fitness effects may alter predicted selection gradients, causing changes in magnitude and sign as the population mean evolves. Second, we show how evolutionary games involving plastic behavioral responses to partners can be modeled using indirect genetic effects, which describe how trait expression changes in response to genes in the social environment. We demonstrate that repeated social interactions in models of reciprocity generate indirect effects and conversely, that estimates of parameters from indirect genetic effect models may be used to predict the evolution of reciprocity. We argue that a pluralistic view incorporating both theoretical approaches will benefit empiricists and theorists studying social evolution. We advocate the measurement of social selection and indirect genetic effects in natural populations to test the predictions from game theory and, in turn, the use of game theory models to aid in the interpretation of quantitative genetic estimates.

Automated cerebral microbleed detection using selective 3D gradient co-occurance matrix and convolutional neural network

Cerebral Microbleeds (CMBs) are small blood clots that form in the brains of many elderly and stroke victims. This promotes dementia as well as other physiological challenges in their daily activities. Early detection of CMBs is critical for limiting the long-term impacts that lead to confusing behavior. Because CMBs in the brain are so minute, manual identification by neurologists might be difficult at times, leading to misinterpretations. In this paper, we propound a newfangled feature extraction technique for the detection of CMB, by Selective 3D Gradient Co-occurance Matrix (S3DGCM). Our proposed framework efficiently detects the cerebral microbleeds (CMB) from magnetic resonance images by incorporating the Scharr Gradient operator with the concepts of the Co-occurance Matrix. The selected ROI was classified as CMB or non-CMB using a Convolutional Neural Network (CNN) classifier with Long Short Term Memory (LSTM) based on the extracted features from the S3DGCM. When compared with the previous methods, our method can extract more useful features which increase the accuracy rate in the detection process. The proposed work has been verified on two datasets containing MRI images with CMBs and Non-CMBs. The experimental results specify that our proposed scheme yields a promising result with the best sensitivity of 98.76%, specificity of 97.21%, an accuracy of 98.24%, and an F1-score of 98.76%.

Individual variation in reaction norms but no directional selection in reproductive plasticity of a wild passerine population.

In the plant–insect–insectivorous bird food chain, directional changes in climate can result in mismatched phenology, potentially affecting selection pressures. Phenotypic plasticity in the timing of breeding, characterized by reaction norm slopes, can help maximize fitness when faced with earlier prey emergence. In temperate passerines, the timing of tree budburst influences food availability for chicks through caterpillar phenology and the resulting food abundance patterns. Thus, the timing of tree budburst might serve as a more direct proxy for the cue to time egg‐laying. The evolutionary potential of breeding plasticity relies on heritable variation, which is based upon individual variation, yet studies on individual variation in plasticity are few. Here, we tested for the laying date—budburst date and the clutch size—laying date reaction norms, and examined 1) the among‐individual variance in reaction norm intercepts and slopes; and 2) the selection differentials and gradients on these intercepts and slopes. Using long‐term data of oak (genus Quercus) budburst and blue tit (Cyanistes caeruleus) reproduction, we applied within‐subject centering to detect reaction norms, followed by bivariate random regression to quantify among‐individual variance in reaction norm properties and their covariance with fitness. Individuals significantly differed in intercepts and slopes of both laying date—budburst date and clutch size—laying date reaction norms, and directional selection was present for an earlier laying date and a larger clutch size (intercepts), but not on plasticity (slopes). We found that individuals have their own regimes for adjusting egg‐laying and clutch size. This study provides further support of individual variation of phenotypic plasticity in birds.

OP29 Peripheral blood DNA methylation biomarkers accurately predict clinical- and endoscopic response to vedolizumab in a real-life cohort of Crohn’s Disease patients

Abstract Background Despite the proven efficacy of vedolizumab (VDZ), only 29% and 36% of the Crohn’s disease (CD) patients present corticosteroid-free clinical- and endoscopic remission, respectively. Therefore, predictive biomarkers for treatment success would be of extreme value. Previous studies have identified aberrant DNA methylation associated with CD-specific phenotypes, suggesting that the methylome may be useful for classification and prediction of VDZ treatment response. Here, we sought to identify such DNA methylation biomarkers that can predict clinical- and endoscopic response to VDZ in CD patients. Methods We prospectively recruited adult CD patients that initiated VDZ treatment following a baseline colonoscopy in two cohorts: a discovery and validation cohort. Peripheral blood DNA methylation profiles were measured prior to treatment (T1), and after a median of 22 weeks (T2) using the Illumina Infinium HumanMethylation EPIC BeadChip array. Response (R) was defined as the strict combination of endoscopic- (≥50% reduction in SES-CD score) and steroid-free clinical response (≥3 point drop in HBI and HBI ≤4 AND no systemic steroids) and/or biochemical response (≥50% reduction in C-reactive protein (CRP) and fecal calprotectin or a CRP ≤5 g/mL and fecal calprotectin ≤250 µg/g). Twenty-one patients had deep remission (DR), defined as a combined endoscopic- (SES-CD≤2) and steroid-free clinical remission (HBI ≤4, no systemic steroids). Biomarker identification and classification analyses were performed using stability selection gradient boosting. Results In total, 64 CD patients were enrolled (discovery 16R/14NR and validation 20R/14NR). Both cohorts were comparable for age, sex and smoking status. Forty-nine (77%) patients had previously failed an anti-TNF agent. All patients had measurable serum vedolizumab concentration at T2 (median 14.5 (6.9 – 21.3) µg/mL). Through classification analysis at T1, we were capable of discriminating R from NR with high predictive performance (25 CpGs, AUC 0.88, F1-score of 0.84, PPV of 0.91 and NPV of 0.85). When analysing the methylome of patients in deep remission, we identified 23 CpGs with high predictive performance upon independent validation (F1-score 0.80, PPV of 0.71 and NPV of 0.86). Investigating the CpGs of interest implicated genes involved in endothelial cell-cell adhesion and integrin dependent T-cell homing, corroborating VDZ’s mode of action. Conclusion We demonstrate two novel 25- and 23-feature panels of epigenetic biomarkers that accurately predict response or deep remission to vedolizumab respectively. Similar analyses on infliximab, adalimumab and ustekinumab are currently ongoing as part of the EPIC-CD study.

Morphological Evolution: Bioinspired Methods for Analyzing Bioinspired Robots.

To fully understand the evolution of complex morphologies, analyses cannot stop at selection: It is essential to investigate the roles and interactions of multiple processes that drive evolutionary outcomes. The challenges of undertaking such analyses have affected both evolutionary biologists and evolutionary roboticists, with their common interests in complex morphologies. In this paper, we present analytical techniques from evolutionary biology, selection gradient analysis and morphospace walks, and we demonstrate their applicability to robot morphologies in analyses of three evolutionary mechanisms: randomness (genetic mutation), development (an explicitly implemented genotype-to-phenotype map), and selection. In particular, we applied these analytical techniques to evolved populations of simulated biorobots—embodied robots designed specifically as models of biological systems, for the testing of biological hypotheses—and we present a variety of results, including analyses that do all of the following: illuminate different evolutionary dynamics for different classes of morphological traits; illustrate how the traits targeted by selection can vary based on the likelihood of random genetic mutation; demonstrate that selection on two selected sets of morphological traits only partially explains the variance in fitness in our biorobots; and suggest that biases in developmental processes could partially explain evolutionary dynamics of morphology. When combined, the complementary analytical approaches discussed in this paper can enable insight into evolutionary processes beyond selection and thereby deepen our understanding of the evolution of robotic morphologies.

XGB-RF: A Hybrid Machine Learning Approach for IoT Intrusion Detection

In the past few years, Internet of Things (IoT) devices have evolved faster and the use of these devices is exceedingly increasing to make our daily activities easier than ever. However, numerous security flaws persist on IoT devices due to the fact that the majority of them lack the memory and computing resources necessary for adequate security operations. As a result, IoT devices are affected by a variety of attacks. A single attack on network systems or devices can lead to significant damages in data security and privacy. However, machine-learning techniques can be applied to detect IoT attacks. In this paper, a hybrid machine learning scheme called XGB-RF is proposed for detecting intrusion attacks. The proposed hybrid method was applied to the N-BaIoT dataset containing hazardous botnet attacks. Random forest (RF) was used for the feature selection and eXtreme Gradient Boosting (XGB) classifier was used to detect different types of attacks on IoT environments. The performance of the proposed XGB-RF scheme is evaluated based on several evaluation metrics and demonstrates that the model successfully detects 99.94% of the attacks. After comparing it with state-of-the-art algorithms, our proposed model has achieved better performance for every metric. As the proposed scheme is capable of detecting botnet attacks effectively, it can significantly contribute to reducing the security concerns associated with IoT systems.