Andrej Aderhold

Performance evaluation of artificial bee colony optimization and new selection schemes

The artificial bee colony optimization (ABC) is a population-based algorithm for function optimization that is inspired by the foraging behavior of bees. The population consists of two types of artificial bees: employed bees (EBs) which scout for new, good solutions and onlooker bees (OBs) that search in the neighborhood of solutions found by the EBs. In this paper we study in detail the influence of ABC’s parameters on its optimization behavior. It is also investigated whether the use of OBs is always advantageous. Moreover, we propose two new variants of ABC which use new methods for the position update of the artificial bees. Extensive empirical tests were performed to compare the new variants with the standard ABC and several other metaheuristics on a set of benchmark functions. Our findings show that the ideal parameter values depend on the hardness of the optimization goal and that the standard values suggested in the literature should be applied with care. Moreover, it is shown that in some situations it is advantageous to use OBs but in others it is not. In addition, a potential problem of the ABC is identified, namely that it performs worse on many functions when the optimum is not located at the center of the search space. Finally it is shown that the new ABC variants improve the algorithm’s performance and achieve very good performance in comparison to other metaheuristics under standard as well as hard optimization goals.

Hierarchical Bayesian models in ecology: Reconstructing species interaction networks from non-homogeneous species abundance data

article i nfo The relationships among organisms and their surroundings can be of immense complexity. To describe and understand an ecosystem as a tangled bank, multiple ways of interaction and their effects have to be considered, suchaspredation,competition,mutualismandfacilitation.Understandingtheresultinginteractionnetworksisa challenge in changing environments, e.g. to predict knock-on effects of invasive species and to understand how climate change impacts biodiversity. The elucidation of complex ecological systems with their interactions will benefit enormously from the development of new machine learning tools that aim to infer the structure of inter- actionnetworks from fielddata.Inthe presentstudy,wepropose a novel Bayesian regressionand multiplechan- gepoint model (BRAM) for reconstructing species interaction networks from observed species distributions. The model has been devised to allow robust inference in the presence of spatial autocorrelation and distributional heterogeneity. We have evaluated the model on simulated data that combines a trophic niche model with a stochastic population model on a 2-dimensional lattice, and we have compared the performance of our model with L1-penalized sparse regression (LASSO) and non-linear Bayesian networks with the BDe scoring scheme. In addition, we have applied our method to plant ground coverage data from the western shore of the Outer Hebrides with the objective to infer the ecological interactions.

Artificial Bee Colony Optimization: A New Selection Scheme and Its Performance

The artificial bee colony optimization (ABC) is a population based algorithm for function optimization that is inspired by the foraging behaviour of bees. The population consists of two types of artificial bees: employed bees (EBs) which scout for new good solution in the search space and onlooker bees (OBs) that search in the neighbourhood of solutions found by the EBs. In this paper we study the influence of the populations size on the optimization behaviour of ABC. Moreover, we investigate when it is advantageous to use OBs. We also propose two variants of ABC which use new methods for the position update of the artificial bees. Empirical tests were performed on a set of benchmark functions. Our findings show that the ideal population size and whether it is advantageous to use OBs depends on the hardness of the optimization goal. Additionally the newly proposed variants of the ABC outperform the standard ABC significantly on all test functions. In comparison to several other optimization algorithm the best ABC variant performs better or at least as good as all reference algorithms in most cases.

Statistical inference of regulatory networks for circadian regulation

Abstract We assess the accuracy of various state-of-the-art statistics and machine learning methods for reconstructing gene and protein regulatory networks in the context of circadian regulation. Our study draws on the increasing availability of gene expression and protein concentration time series for key circadian clock components in Arabidopsis thaliana. In addition, gene expression and protein concentration time series are simulated from a recently published regulatory network of the circadian clock in A. thaliana, in which protein and gene interactions are described by a Markov jump process based on Michaelis-Menten kinetics. We closely follow recent experimental protocols, including the entrainment of seedlings to different light-dark cycles and the knock-out of various key regulatory genes. Our study provides relative network reconstruction accuracy scores for a critical comparative performance evaluation, and sheds light on a series of highly relevant questions: it quantifies the influence of systematically missing values related to unknown protein concentrations and mRNA transcription rates, it investigates the dependence of the performance on the network topology and the degree of recurrency, it provides deeper insight into when and why non-linear methods fail to outperform linear ones, it offers improved guidelines on parameter settings in different inference procedures, and it suggests new hypotheses about the structure of the central circadian gene regulatory network in A. thaliana.

Approximate Bayesian inference in semi-mechanistic models

Inference of interaction networks represented by systems of differential equations is a challenging problem in many scientific disciplines. In the present article, we follow a semi-mechanistic modelling approach based on gradient matching. We investigate the extent to which key factors, including the kinetic model, statistical formulation and numerical methods, impact upon performance at network reconstruction. We emphasize general lessons for computational statisticians when faced with the challenge of model selection, and we assess the accuracy of various alternative paradigms, including recent widely applicable information criteria and different numerical procedures for approximating Bayes factors. We conduct the comparative evaluation with a novel inferential pipeline that systematically disambiguates confounding factors via an ANOVA scheme.

Changes and classification in myocardial contractile function in the left ventricle following acute myocardial infarction

In this research, we hypothesized that novel biomechanical parameters are discriminative in patients following acute ST-segment elevation myocardial infarction (STEMI). To identify these biomechanical biomarkers and bring computational biomechanics ‘closer to the clinic’, we applied state-of-the-art multiphysics cardiac modelling combined with advanced machine learning and multivariate statistical inference to a clinical database of myocardial infarction. We obtained data from 11 STEMI patients (ClinicalTrials.gov NCT01717573) and 27 healthy volunteers, and developed personalized mathematical models for the left ventricle (LV) using an immersed boundary method. Subject-specific constitutive parameters were achieved by matching to clinical measurements. We have shown, for the first time, that compared with healthy controls, patients with STEMI exhibited increased LV wall active tension when normalized by systolic blood pressure, which suggests an increased demand on the contractile reserve of remote functional myocardium. The statistical analysis reveals that the required patient-specific contractility, normalized active tension and the systolic myofilament kinematics have the strongest explanatory power for identifying the myocardial function changes post-MI. We further observed a strong correlation between two biomarkers and the changes in LV ejection fraction at six months from baseline (the required contractility (r = − 0.79, p < 0.01) and the systolic myofilament kinematics (r = 0.70, p = 0.02)). The clinical and prognostic significance of these biomechanical parameters merits further scrutinization.

Inferring bi-directional interactions between circadian clock genes and metabolism with model ensembles

Abstract There has been much interest in reconstructing bi-directional regulatory networks linking the circadian clock to metabolism in plants. A variety of reverse engineering methods from machine learning and computational statistics have been proposed and evaluated. The emphasis of the present paper is on combining models in a model ensemble to boost the network reconstruction accuracy, and to explore various model combination strategies to maximize the improvement. Our results demonstrate that a rich ensemble of predictors outperforms the best individual model, even if the ensemble includes poor predictors with inferior individual reconstruction accuracy. For our application to metabolomic and transcriptomic time series from various mutagenesis plants grown in different light-dark cycles we also show how to determine the optimal time lag between interactions, and we identify significant interactions with a randomization test. Our study predicts new statistically significant interactions between circadian clock genes and metabolites in Arabidopsis thaliana, and thus provides independent statistical evidence that the regulation of metabolism by the circadian clock is not uni-directional, but that there is a statistically significant feedback mechanism aiming from metabolism back to the circadian clock.

Targeting Bayes factors with direct-path non-equilibrium thermodynamic integration

Thermodynamic integration (TI) for computing marginal likelihoods is based on an inverse annealing path from the prior to the posterior distribution. In many cases, the resulting estimator suffers from high variability, which particularly stems from the prior regime. When comparing complex models with differences in a comparatively small number of parameters, intrinsic errors from sampling fluctuations may outweigh the differences in the log marginal likelihood estimates. In the present article, we propose a TI scheme that directly targets the log Bayes factor. The method is based on a modified annealing path between the posterior distributions of the two models compared, which systematically avoids the high variance prior regime. We combine this scheme with the concept of non-equilibrium TI to minimise discretisation errors from numerical integration. Results obtained on Bayesian regression models applied to standard benchmark data, and a complex hierarchical model applied to biopathway inference, demonstrate a significant reduction in estimator variance over state-of-the-art TI methods.