Long-term stimulation by implanted pacemaker enables non-atrophic treatment of bilateral vocal fold paresis in a human-like animal model.

A wide variety of treatments have been developed to improve respiratory function and quality of life in patients with bilateral vocal fold paresis (BVFP). One experimental method is the electrical activation of the posterior cricoarytenoid (PCA) muscle with a laryngeal pacemaker (LP) to open the vocal folds. We used an ovine (sheep) model of unilateral VFP to study the long-term effects of functional electrical stimulation on the PCA muscles. The left recurrent laryngeal nerve was cryo-damaged in all animals and an LP was implanted except for the controls. After a reinnervation phase of six months, animals were pooled into groups that received either no treatment, implantation of an LP only, or implantation of an LP and six months of stimulation with different duty cycles. Automated image analysis of fluorescently stained PCA cross-sections was performed to assess relevant muscle characteristics. We observed a fast-to-slow fibre type shift in response to nerve damage and stimulation, but no complete conversion to a slow-twitch-muscle. Fibre size, proportion of hybrid fibres, and intramuscular collagen content were not substantially altered by the stimulation. These results demonstrate that 30 Hz burst stimulation with duty cycles of 40% and 70% did not induce PCA atrophy or fibrosis. Thus, long-term stimulation with an LP is a promising approach for treating BVFP in humans without compromising muscle conditions.

Quantitative analysis of peroxisome tracks using a Hidden Markov Model.

Diffusion and mobility are essential for cellular functions, as molecules are usually distributed throughout the cell and have to meet to interact and perform their function. This also involves the cytosolic migration of cellular organelles. However, observing such diffusion and interaction dynamics is challenging due to the high spatial and temporal resolution required and the accurate analysis of the diffusional tracks. The latter is especially important when identifying anomalous diffusion events, such as directed motions, which are often rare. Here, we investigate the migration modes of peroxisome organelles in the cytosol of living cells. Peroxisomes predominantly migrate randomly, but occasionally they bind to the cell's microtubular network and perform directed migration, which is difficult to quantify, and so far, accurate analysis of switching between these migration modes is missing. We set out to solve this limitation by experiments and analysis with high statistical accuracy. Specifically, we collect temporal diffusion tracks of thousands of individual peroxisomes in the HEK 293 cell line using two-dimensional spinning disc fluorescence microscopy at a high acquisition rate of 10 frames/s. We use a Hidden Markov Model with two hidden states to (1) automatically identify directed migration segments of the tracks and (2) quantify the migration properties for comparison between states and between different experimental conditions. Comparing different cellular conditions, we show that the knockout of the peroxisomal membrane protein PEX14 leads to a decrease in the directed movement due to a lowered binding probability to the microtubule. However, it does not eradicate binding, highlighting further microtubule-binding mechanisms of peroxisomes than via PEX14. In contrast, structural changes of the microtubular network explain perceived eradication of directed movement by disassembly of microtubules by Nocodazole-treatment.

IMFSegNet: Cost-effective and objective quantification of intramuscular fat in histological sections by deep learning.

The assessment of muscle condition is of great importance in various research areas. In particular, evaluating the degree of intramuscular fat (IMF) in tissue sections is a challenging task, which today is still mostly performed qualitatively or quantitatively by a highly subjective and error-prone manual analysis. We here realize the mission to make automated IMF analysis possible that (i) minimizes subjectivity, (ii) provides accurate and quantitative results quickly, and (iii) is cost-effective using standard hematoxylin and eosin (H&E) stained tissue sections. To address all these needs in a deep learning approach, we utilized the convolutional encoder-decoder network SegNet to train the specialized network IMFSegNet allowing to accurately quantify the spatial distribution of IMF in histological sections. Our fully automated analysis was validated on 17 H&E-stained muscle sections from individual sheep and compared to various state-of-the-art approaches. Not only does IMFSegNet outperform all other approaches, but this neural network also provides fully automated and highly accurate results utilizing the most cost-effective procedures of sample preparation and imaging. Furthermore, we shed light on the opacity of black-box approaches such as neural networks by applying an explainable artificial intelligence technique to clarify that the success of IMFSegNet actually lies in identifying the hard-to-detect IMF structures. Embedded in our open-source visual programming language JIPipe that does not require programming skills, it can be expected that IMFSegNet advances muscle condition assessment in basic research across multiple areas as well as in research fields focusing on translational clinical applications.

Targeting of phagolysosomes containing conidia of the fungus Aspergillus fumigatus with polymeric particles.

Conidia of the airborne human-pathogenic fungus Aspergillus fumigatus are inhaled by humans. In the lung, they are phagocytosed by alveolar macrophages and intracellularly processed. In macrophages, however, conidia can interfere with the maturation of phagolysosomes to avoid their elimination. To investigate whether polymeric particles (PPs) can reach this intracellular pathogen in macrophages, we formulated dye-labeled PPs with a size allowing for their phagocytosis. PPs were efficiently taken up by RAW 264.7 macrophages and were found in phagolysosomes. When macrophages were infected with conidia prior to the addition of PPs, we found that they co-localized in the same phagolysosomes. Mechanistically, the fusion of phagolysosomes containing PPs with phagolysosomes containing conidia was observed. Increasing concentrations of PPs increased fusion events, resulting in 14% of phagolysosomes containing both conidia and PPs. We demonstrate that PPs can reach conidia-containing phagolysosomes, making these particles a promising carrier system for antimicrobial drugs to target intracellular pathogens. KEY POINTS: • Polymer particles of a size larger than 500 nm are internalized by macrophages and localized in phagolysosomes. • These particles can be delivered to Aspergillus fumigatus conidia-containing phagolysosomes of macrophages. • Enhanced phagolysosome fusion by the use of vacuolin1 can increase particle delivery.

Human Neutrophils Produce Antifungal Extracellular Vesicles against Aspergillus fumigatus.

Polymorphonuclear granulocytes (PMNs) are indispensable for controlling life-threatening fungal infections. In addition to various effector mechanisms, PMNs also produce extracellular vesicles (EVs). Their contribution to antifungal defense has remained unexplored. We reveal that the clinically important human-pathogenic fungus Aspergillus fumigatus triggers PMNs to release a distinct set of antifungal EVs (afEVs). Proteome analyses indicated that afEVs are enriched in antimicrobial proteins. The cargo and the release kinetics of EVs are modulated by the fungal strain confronted. Tracking of afEVs indicated that they associated with fungal cells and even entered fungal hyphae, resulting in alterations in the morphology of the fungal cell wall and dose-dependent antifungal effects. To assess as a proof of concept whether the antimicrobial proteins found in afEVs might contribute to growth inhibition of hyphae when present in the fungal cytoplasm, two human proteins enriched in afEVs, cathepsin G and azurocidin, were heterologously expressed in fungal hyphae. This led to reduced fungal growth relative to that of a control strain producing the human retinol binding protein 7. In conclusion, extracellular vesicles produced by neutrophils in response to A. fumigatus infection are able to associate with the fungus, limit growth, and elicit cell damage by delivering antifungal cargo. This finding offers an intriguing, previously overlooked mechanism of antifungal defense against A. fumigatusIMPORTANCE Invasive fungal infections caused by the mold Aspergillus fumigatus are a growing concern in the clinic due to the increasing use of immunosuppressive therapies and increasing antifungal drug resistance. These infections result in high rates of mortality, as treatment and diagnostic options remain limited. In healthy individuals, neutrophilic granulocytes are critical for elimination of A. fumigatus from the host; however, the exact extracellular mechanism of neutrophil-mediated antifungal activity remains unresolved. Here, we present a mode of antifungal defense employed by human neutrophils against A. fumigatus not previously described. We found that extracellular vesicles produced by neutrophils in response to A. fumigatus infection are able to associate with the fungus, limit growth, and elicit cell damage by delivering antifungal cargo. In the end, antifungal extracellular vesicle biology provides a significant step forward in our understanding of A. fumigatus host pathogenesis and opens up novel diagnostic and therapeutic possibilities.

Regulation of the Germinal Center Reaction and Somatic Hypermutation Dynamics by Homologous Recombination.

During somatic hypermutation (SHM) of Ig genes in germinal center B cells, lesions introduced by activation-induced cytidine deaminase are processed by multiple error-prone repair pathways. Although error-free repair by homologous recombination (HR) is crucial to prevent excessive DNA strand breakage at activation-induced cytidine deaminase off-target genes, its role at the hypermutating Ig locus in the germinal center is unexplored. Using B cell-specific inactivation of the critical HR factor Brca2, we detected decreased proliferation, survival, and thereby class switching of ex vivo-activated B cells. Intriguingly, an HR defect allowed for a germinal center reaction and affinity maturation in vivo, albeit at reduced amounts. Analysis of SHM revealed that a certain fraction of DNA lesions at C:G bp was indeed repaired in an error-free manner via Brca2 instead of being processed by error-prone translesion polymerases. By applying a novel pseudo-time in silico analysis of mutational processes, we found that the activity of A:T mutagenesis during SHM increased during a germinal center reaction, but this was in part defective in Brca2-deficient mice. These mutation pattern changes in Brca2-deficient B cells were mostly specific for the Ig V region, suggesting a local or time-dependent need for recombination repair to survive high rates of SHM and especially A:T mutagenesis.

Lack of CD45 in FLT3-ITD mice results in a myeloproliferative phenotype, cortical porosity, and ectopic bone formation.

The receptor tyrosine kinase FLT3 is expressed in myeloid and lymphoid progenitor cells. Activating mutations in FLT3 occur in 25-30% of acute myeloid leukaemia (AML) patients. Most common are internal tandem duplications of sequence (ITD) leading to constitutive FLT3-ITD kinase activity with an altered signalling quality promoting leukaemic cell transformation. Here, we observed the attenuating role of the receptor-like protein tyrosine phosphatase (RPTP) CD45/Ptprc in FLT3 signalling in vivo. Low level expression of this abundant RPTP correlates with a poor prognosis of FLT3-ITD-positive AML patients. To get a further insight into the regulatory role of Ptprc in FLT3-ITD activity in vivo, Ptprc knock-out mice were bred with FLT3-ITD knock-in mice. Inactivation of the Ptprc gene in FLT3-ITD mice resulted in a drastically shortened life span and development of severe monocytosis, a block in B-cell development and anaemia. The myeloproliferative phenotype was associated with extramedullary haematopoiesis, splenohepatomegaly and severe alterations of organ structures. The phenotypic alterations were associated with increased transforming signalling of FLT3-ITD, including activation of its downstream target STAT5. These data reveal the capacity of Ptprc for the regulation of FLT3-ITD signalling activity in vivo. In addition, histopathology and computed tomography (CT) revealed an unexpected bone phenotype; the FLT3-ITD Ptprc-/- mice, but none of the controls, showed pronounced alterations in bone morphology and, in part, apparent features of osteoporosis. In the spleen, ectopic bone formation was observed. The observed bone phenotypes suggest a previously unappreciated capacity of FLT3-ITD (and presumably FLT3) to regulate bone development/remodelling, which is under negative control of CD45/Ptprc.

Clinical S. aureus Isolates Vary in Their Virulence to Promote Adaptation to the Host.

Staphylococcus aureus colonizes epithelial surfaces, but it can also cause severe infections. The aim of this work was to investigate whether bacterial virulence correlates with defined types of tissue infections. For this, we collected 10⁻12 clinical S. aureus strains each from nasal colonization, and from patients with endoprosthesis infection, hematogenous osteomyelitis, and sepsis. All strains were characterized by genotypic analysis, and by the expression of virulence factors. The host⁻pathogen interaction was studied through several functional assays in osteoblast cultures. Additionally, selected strains were tested in a murine sepsis/osteomyelitis model. We did not find characteristic bacterial features for the defined infection types; rather, a wide range in all strain collections regarding cytotoxicity and invasiveness was observed. Interestingly, all strains were able to persist and to form small colony variants (SCVs). However, the low-cytotoxicity strains survived in higher numbers, and were less efficiently cleared by the host than the highly cytotoxic strains. In summary, our results indicate that not only destructive, but also low-cytotoxicity strains are able to induce infections. The low-cytotoxicity strains can successfully survive, and are less efficiently cleared from the host than the highly cytotoxic strains, which represent a source for chronic infections. The understanding of this interplay/evolution between the host and the pathogen during infection, with specific attention towards low-cytotoxicity isolates, will help to optimize treatment strategies for invasive and therapy-refractory infection courses.

Coding of Experimental Conditions in Microfluidic Droplet Assays Using Colored Beads and Machine Learning Supported Image Analysis.

To efficiently exploit the potential of several millions of droplets that can be considered as individual bioreactors in microfluidic experiments, methods to encode different experimental conditions in droplets are needed. The approach presented here is based on coencapsulation of colored polystyrene beads with biological samples. The decoding of the droplets, as well as content quantification, are performed by automated analysis of triggered images of individual droplets in-flow using bright-field microscopy. The decoding strategy combines bead classification using a random forest classifier and Bayesian inference to identify different codes and thus experimental conditions. Antibiotic susceptibility testing of nine different antibiotics and the determination of the minimal inhibitory concentration of a specific antibiotic against a laboratory strain of Escherichia coli are presented as a proof-of-principle. It is demonstrated that this method allows successful encoding and decoding of 20 different experimental conditions within a large droplet population of more than 105 droplets per condition. The decoding strategy correctly assigns 99.6% of droplets to the correct condition and a method for the determination of minimal inhibitory concentration using droplet microfluidics is established. The current encoding and decoding pipeline can readily be extended to more codes by adding more bead colors or color combinations.

Candida albicans-Induced Epithelial Damage Mediates Translocation through Intestinal Barriers.

Life-threatening systemic infections often occur due to the translocation of pathogens across the gut barrier and into the bloodstream. While the microbial and host mechanisms permitting bacterial gut translocation are well characterized, these mechanisms are still unclear for fungal pathogens such as Candida albicans, a leading cause of nosocomial fungal bloodstream infections. In this study, we dissected the cellular mechanisms of translocation of C. albicans across intestinal epithelia in vitro and identified fungal genes associated with this process. We show that fungal translocation is a dynamic process initiated by invasion and followed by cellular damage and loss of epithelial integrity. A screen of >2,000 C. albicans deletion mutants identified genes required for cellular damage of and translocation across enterocytes. Correlation analysis suggests that hypha formation, barrier damage above a minimum threshold level, and a decreased epithelial integrity are required for efficient fungal translocation. Translocation occurs predominantly via a transcellular route, which is associated with fungus-induced necrotic epithelial damage, but not apoptotic cell death. The cytolytic peptide toxin of C. albicans, candidalysin, was found to be essential for damage of enterocytes and was a key factor in subsequent fungal translocation, suggesting that transcellular translocation of C. albicans through intestinal layers is mediated by candidalysin. However, fungal invasion and low-level translocation can also occur via non-transcellular routes in a candidalysin-independent manner. This is the first study showing translocation of a human-pathogenic fungus across the intestinal barrier being mediated by a peptide toxin.IMPORTANCECandida albicans, usually a harmless fungus colonizing human mucosae, can cause lethal bloodstream infections when it manages to translocate across the intestinal epithelium. This can result from antibiotic treatment, immune dysfunction, or intestinal damage (e.g., during surgery). However, fungal processes may also contribute. In this study, we investigated the translocation process of C. albicans using in vitro cell culture models. Translocation occurs as a stepwise process starting with invasion, followed by epithelial damage and loss of epithelial integrity. The ability to secrete candidalysin, a peptide toxin deriving from the hyphal protein Ece1, is key: C. albicans hyphae, secreting candidalysin, take advantage of a necrotic weakened epithelium to translocate through the intestinal layer.

Untangling cell tracks: Quantifying cell migration by time lapse image data analysis.

Automated microscopy has given researchers access to great amounts of live cell imaging data from in vitro and in vivo experiments. Much focus has been put on extracting cell tracks from such data using a plethora of segmentation and tracking algorithms, but further analysis is normally required to draw biologically relevant conclusions. Such relevant conclusions may be whether the migration is directed or not, whether the population has homogeneous or heterogeneous migration patterns. This review focuses on the analysis of cell migration data that are extracted from time lapse images. We discuss a range of measures and models used to analyze cell tracks independent of the biological system or the way the tracks were obtained. For single-cell migration, we focus on measures and models giving examples of biological systems where they have been applied, for example, migration of bacteria, fibroblasts, and immune cells. For collective migration, we describe the model systems wound healing, neural crest migration, and Drosophila gastrulation and discuss methods for cell migration within these systems. We also discuss the role of the extracellular matrix and subsequent differences between track analysis in vitro and in vivo. Besides methods and measures, we are putting special focus on the need for openly available data and code, as well as a lack of common vocabulary in cell track analysis. © 2017 International Society for Advancement of Cytometry.

Mast cells acquire MHCII from dendritic cells during skin inflammation.

Mast cells (MCs) and dendritic cells (DCs) are essential innate sentinels populating host-environment interfaces. Using longitudinal intravital multiphoton microscopy of DCGFP/MCRFP reporter mice, we herein provide in vivo evidence that migratory DCs execute targeted cell-to-cell interactions with stationary MCs before leaving the inflamed skin to draining lymph nodes. During initial stages of skin inflammation, DCs dynamically scan MCs, whereas at a later stage, long-lasting interactions predominate. These innate-to-innate synapse-like contacts ultimately culminate in DC-to-MC molecule transfers including major histocompatibility complex class II (MHCII) proteins enabling subsequent ex vivo priming of allogeneic T cells with a specific cytokine signature. The extent of MHCII transfer to MCs correlates with their T cell priming efficiency. Importantly, preventing the cross talk by preceding DC depletion decreases MC antigen presenting capacity and T cell-driven inflammation. Consequently, we identify an innate intercellular communication arming resident MCs with key DC functions that might contribute to the acute defense potential during critical periods of migration-based DC absence.

Automated Quantification of Early Bone Alterations and Pathological Bone Turnover in Experimental Arthritis by in vivo PET/CT Imaging.

The assessment of bone damage is required to evaluate disease severity and treatment efficacy both in arthritis patients and in experimental arthritis models. Today there is still a lack of in vivo methods that enable the quantification of arthritic processes at an early stage of the disease. We performed longitudinal in vivo imaging with [18F]-fluoride PET/CT before and after experimental arthritis onset for diseased and control DBA/1 mice and assessed arthritis progression by clinical scoring, tracer uptake studies and bone volume as well as surface roughness measurements. Arthritic animals showed significantly increased tracer uptake in the paws compared to non-diseased controls. Automated CT image analysis revealed increased bone surface roughness already in the earliest stage of the disease. Moreover, we observed clear differences between endosteal and periosteal sites of cortical bone regarding surface roughness. This study shows that in vivo PET/CT imaging is a favorable method to study arthritic processes, enabling the quantification of different aspects of the disease like pathological bone turnover and bone alteration. Especially the evaluation of bone surface roughness is sensitive to early pathological changes and can be applied to study the dynamics of bone erosion at different sites of the bones in an automated fashion.

Quantification of arthritic bone degradation by analysis of 3D micro-computed tomography data.

The use of animal models of arthritis is a key component in the evaluation of therapeutic strategies against the human disease rheumatoid arthritis (RA). Here we present quantitative measurements of bone degradation characterised by the cortical bone profile using glucose-6-phosphate isomerase (G6PI) induced arthritis. We applied micro-computed tomography (µCT) during three arthritis experiments and one control experiment to image the metatarsals of the hind paws and to investigate the effect of experimental arthritis on their cortical bone profile. For measurements of the cortical profile we automatically identified slices that are orthogonal to individual metatarsals, thereby making the measurements independent of animal placement in the scanner. We measured the average cortical thickness index (CTI) of the metatarsals, as well as the thickness changes along the metatarsal. In this study we introduced the cortical thickness gradient (CTG) as a new measure and we investigated how arthritis affects this measure. We found that in general both CTI and CTG are able to quantify arthritic progression, whilst CTG was found to be the more sensitive measure.

Automated Classification of Circulating Tumor Cells and the Impact of Interobsever Variability on Classifier Training and Performance.

Application of personalized medicine requires integration of different data to determine each patient's unique clinical constitution. The automated analysis of medical data is a growing field where different machine learning techniques are used to minimize the time-consuming task of manual analysis. The evaluation, and often training, of automated classifiers requires manually labelled data as ground truth. In many cases such labelling is not perfect, either because of the data being ambiguous even for a trained expert or because of mistakes. Here we investigated the interobserver variability of image data comprising fluorescently stained circulating tumor cells and its effect on the performance of two automated classifiers, a random forest and a support vector machine. We found that uncertainty in annotation between observers limited the performance of the automated classifiers, especially when it was included in the test set on which classifier performance was measured. The random forest classifier turned out to be resilient to uncertainty in the training data while the support vector machine's performance is highly dependent on the amount of uncertainty in the training data. We finally introduced the consensus data set as a possible solution for evaluation of automated classifiers that minimizes the penalty of interobserver variability.

Automated quantification of the phagocytosis of Aspergillus fumigatus conidia by a novel image analysis algorithm.

Studying the pathobiology of the fungus Aspergillus fumigatus has gained a lot of attention in recent years. This is due to the fact that this fungus is a human pathogen that can cause severe diseases, like invasive pulmonary aspergillosis in immunocompromised patients. Because alveolar macrophages belong to the first line of defense against the fungus, here, we conduct an image-based study on the host-pathogen interaction between murine alveolar macrophages and A. fumigatus. This is achieved by an automated image analysis approach that uses a combination of thresholding, watershed segmentation and feature-based object classification. In contrast to previous approaches, our algorithm allows for the segmentation of individual macrophages in the images and this enables us to compute the distribution of phagocytosed and macrophage-adherent conidia over all macrophages. The novel automated image-based analysis provides access to all cell-cell interactions in the assay and thereby represents a framework that enables comprehensive computation of diverse characteristic parameters and comparative investigation for different strains. We here apply automated image analysis to confocal laser scanning microscopy images of the two wild-type strains ATCC 46645 and CEA10 of A. fumigatus and investigate the ability of macrophages to phagocytose the respective conidia. It is found that the CEA10 strain triggers a stronger response of the macrophages as revealed by a higher phagocytosis ratio and a larger portion of the macrophages being active in the phagocytosis process.

Estimation of cortical magnification from positional error in normally sighted and amblyopic subjects.

We describe a method for deriving the linear cortical magnification factor from positional error across the visual field. We compared magnification obtained from this method between normally sighted individuals and amblyopic individuals, who receive atypical visual input during development. The cortical magnification factor was derived for each subject from positional error at 32 locations in the visual field, using an established model of conformal mapping between retinal and cortical coordinates. Magnification of the normally sighted group matched estimates from previous physiological and neuroimaging studies in humans, confirming the validity of the approach. The estimate of magnification for the amblyopic group was significantly lower than the normal group: by 4.4 mm deg(-1) at 1° eccentricity, assuming a constant scaling factor for both groups. These estimates, if correct, suggest a role for early visual experience in establishing retinotopic mapping in cortex. We discuss the implications of altered cortical magnification for cortical size, and consider other neural changes that may account for the amblyopic results.

Automated detection of circulating tumor cells with naive Bayesian classifiers.

Personalized medicine is a modern healthcare approach where information on each person's unique clinical constitution is exploited to realize early disease intervention based on more informed medical decisions. The application of diagnostic tools in combination with measurement evaluation that can be performed in a reliable and automated fashion plays a key role in this context. As the progression of various cancer diseases and the effectiveness of their treatments are related to a varying number of tumor cells that circulate in blood, the determination of their extremely low numbers by liquid biopsy is a decisive prognostic marker. To detect and enumerate circulating tumor cells (CTCs) in a reliable and automated fashion, we apply methods from machine learning using a naive Bayesian classifier (NBC) based on a probabilistic generative mixture model. Cells are collected with a functionalized medical wire and are stained for fluorescence microscopy so that their color signature can be used for classification through the construction of Red-Green-Blue (RGB) color histograms. Exploiting the information on the fluorescence signature of CTCs by the NBC does not only allow going beyond previous approaches but also provides a method of unsupervised learning that is required for unlabeled training data. A quantitative comparison with a state-of-the-art support vector machine, which requires labeled data, demonstrates the competitiveness of the NBC method.

Using evolutionary algorithms for fitting high-dimensional models to neuronal data.

In the study of neurosciences, and of complex biological systems in general, there is frequently a need to fit mathematical models with large numbers of parameters to highly complex datasets. Here we consider algorithms of two different classes, gradient following (GF) methods and evolutionary algorithms (EA) and examine their performance in fitting a 9-parameter model of a filter-based visual neuron to real data recorded from a sample of 107 neurons in macaque primary visual cortex (V1). Although the GF method converged very rapidly on a solution, it was highly susceptible to the effects of local minima in the error surface and produced relatively poor fits unless the initial estimates of the parameters were already very good. Conversely, although the EA required many more iterations of evaluating the model neuron's response to a series of stimuli, it ultimately found better solutions in nearly all cases and its performance was independent of the starting parameters of the model. Thus, although the fitting process was lengthy in terms of processing time, the relative lack of human intervention in the evolutionary algorithm, and its ability ultimately to generate model fits that could be trusted as being close to optimal, made it far superior in this particular application than the gradient following methods. This is likely to be the case in many further complex systems, as are often found in neuroscience.