Public Library of Science RSS Feed
https://plos.figshare.com/
RSS feed for Figshare Institution Public Library of ScienceeyJhbGciOiJIUzI1NiJ9.eyJ2YWx1ZXMiOlsxNjA2ODQ4MTI0MDAwLDEzMzE1MDc2XX0._KAjKnFU7fx1J6D7PjQSVKBjEDOPpJWCUx1-eYUIzbYBiologically-informed neural networks guide mechanistic modeling from sparse experimental data
https://plos.figshare.com/collections/Biologically-informed_neural_networks_guide_mechanistic_modeling_from_sparse_experimental_data/5225183
Biologically-informed neural networks (BINNs), an extension of physics-informed neural networks [1], are introduced and used to discover the underlying dynamics of biological systems from sparse experimental data. In the present work, BINNs are trained in a supervised learning framework to approximate in vitro cell biology assay experiments while respecting a generalized form of the governing reaction-diffusion partial differential equation (PDE). By allowing the diffusion and reaction terms to be multilayer perceptrons (MLPs), the nonlinear forms of these terms can be learned while simultaneously converging to the solution of the governing PDE. Further, the trained MLPs are used to guide the selection of biologically interpretable mechanistic forms of the PDE terms which provides new insights into the biological and physical mechanisms that govern the dynamics of the observed system. The method is evaluated on sparse real-world data from wound healing assays with varying initial cell densities [2].Biophysics, Physiology, Biotechnology, Cancer, Science Policy, Biological Sciences not elsewhere classified, Information Systems not elsewhere classified2020-12-01 18:43:06Difference Akaike Information Criterion (ΔAIC) scores.
https://plos.figshare.com/articles/dataset/Difference_Akaike_Information_Criterion_AIC_scores_/13315172
Difference Akaike Information Criterion (ΔAIC) scores.Biophysics, Physiology, Biotechnology, Cancer, Science Policy, Biological Sciences not elsewhere classified, Information Systems not elsewhere classified2020-12-01 18:43:05Akaike Information Criterion (AIC) scores.
https://plos.figshare.com/articles/dataset/Akaike_Information_Criterion_AIC_scores_/13315169
Akaike Information Criterion (AIC) scores.Biophysics, Physiology, Biotechnology, Cancer, Science Policy, Biological Sciences not elsewhere classified, Information Systems not elsewhere classified2020-12-01 18:43:04Generalized least squares (GLS) errors.
https://plos.figshare.com/articles/dataset/Generalized_least_squares_GLS_errors_/13315166
Generalized least squares (GLS) errors.Biophysics, Physiology, Biotechnology, Cancer, Science Policy, Biological Sciences not elsewhere classified, Information Systems not elsewhere classified2020-12-01 18:43:02BINN-guided delay-reaction-diffusion model parameters.
https://plos.figshare.com/articles/dataset/BINN-guided_delay-reaction-diffusion_model_parameters_/13315160
BINN-guided delay-reaction-diffusion model parameters.Biophysics, Physiology, Biotechnology, Cancer, Science Policy, Biological Sciences not elsewhere classified, Information Systems not elsewhere classified2020-12-01 18:43:01BINN-guided delay-reaction-diffusion model solutions.
https://plos.figshare.com/articles/figure/BINN-guided_delay-reaction-diffusion_model_solutions_/13315157
Predicted cell density profiles using the delay-reaction-diffusion model in Eq (11). Each subplot corresponds to an experiment with a different initial cell density (i.e. 10,000, 12,000, 14,000, 16,000, 18,000, and 20,000 cells per well). Solid lines represent the numerical solution to Eq (11) using the parameters that minimize in Eq (6). The markers represent the experimental scratch assay data.Biophysics, Physiology, Biotechnology, Cancer, Science Policy, Biological Sciences not elsewhere classified, Information Systems not elsewhere classified2020-12-01 18:42:59Delay-reaction-diffusion BINN terms.
https://plos.figshare.com/articles/figure/Delay-reaction-diffusion_BINN_terms_/13315154
The learned diffusivity, DMLP, growth, GMLP, and delay, TMLP, functions extracted from the corresponding BINNs with governing delay-reaction-diffusion PDE in Eq (10). Each line corresponds to an experiment with a different initial cell density (i.e. 10,000, 12,000, 14,000, 16,000, 18,000, and 20,000 cells per well). Note that DMLP and GMLP have different lengths since they are evaluated between the minimum and maximum observed cell densities corresponding to each data set.Biophysics, Physiology, Biotechnology, Cancer, Science Policy, Biological Sciences not elsewhere classified, Information Systems not elsewhere classified2020-12-01 18:42:58Delay-reaction-diffusion BINN terms and discrepancy.
https://plos.figshare.com/articles/figure/Delay-reaction-diffusion_BINN_terms_and_discrepancy_/13315151
Left: learned diffusivity and growth functions, DMLP and GMLP, evaluated over cell density, u, and delay function, TMLP, evaluated over time, t. Right: Predicted cell density profiles using BINNs with the governing delay-reaction-diffusion PDE in Eq (10) for data with initial cell density 20,000 cells per well. Solid lines represent the numerical solution to Eq (10) using DMLP, GMLP, and TMLP. The markers represent the experimental scratch assay data.Biophysics, Physiology, Biotechnology, Cancer, Science Policy, Biological Sciences not elsewhere classified, Information Systems not elsewhere classified2020-12-01 18:42:56Reaction-diffusion BINN terms and discrepancy.
https://plos.figshare.com/articles/figure/Reaction-diffusion_BINN_terms_and_discrepancy_/13315148
Left: learned diffusivity and growth functions, DMLP and GMLP, evaluated over cell density, u. Right: Predicted cell density profiles using BINNs with the governing reaction-diffusion PDE in Eq (9) for data with initial cell density 20,000 cells per well. Solid lines represent the numerical solution to Eq (9) using DMLP and GMLP. The markers represent the experimental scratch assay data.Biophysics, Physiology, Biotechnology, Cancer, Science Policy, Biological Sciences not elsewhere classified, Information Systems not elsewhere classified2020-12-01 18:42:54Biologically-informed neural networks for reaction-diffusion models.
https://plos.figshare.com/articles/figure/Biologically-informed_neural_networks_for_reaction-diffusion_models_/13315145
(A) BINNs are deep neural networks that approximate the solution of a governing dynamical system. (B) By allowing the terms of the dynamical system (e.g. diffusivity function and growth function ) to be function-approximating deep neural networks, the nonlinear forms of these terms can be learned without the need to specify a mechanistic model or library of candidate terms. (C) Automatic differentiation is used on compositions of the different neural network models (e.g. u, D, and G) to construct the PDE that describes the governing dynamical system. (D) The governing system is used in the neural network objective function to jointly learn and satisfy the governing PDE while minimizing the error between the network outputs and noisy observations.Biophysics, Physiology, Biotechnology, Cancer, Science Policy, Biological Sciences not elsewhere classified, Information Systems not elsewhere classified2020-12-01 18:42:53Experimental scratch assay data.
https://plos.figshare.com/articles/figure/Experimental_scratch_assay_data_/13315142
Pre-processed cell density profiles from scratch assay experiments with varying initial cell densities [2]. Each subplot corresponds to an experiment with a different initial cell density (i.e. 10,000, 12,000, 14,000, 16,000, 18,000, and 20,000 cells per well). The cell densities are reported at 37 equally-spaced positions and five equally-spaced time points.Biophysics, Physiology, Biotechnology, Cancer, Science Policy, Biological Sciences not elsewhere classified, Information Systems not elsewhere classified2020-12-01 18:42:52Scratch assay experiment.
https://plos.figshare.com/articles/figure/Scratch_assay_experiment_/13315139
(a) An illustration of an experiment with the IncuCyte ZOOM system (Essen BioScience, MI USA). Full details of the experiment and image processing can be found in [2]. Cells are seeded uniformly within each well in a 96-well plate at a pre-specified density of between 10,000 and 20,000 cells per well. A WoundMaker (Essen BioScience) is used to create a uniform vertical scratch along the middle of the well. (b) Microscopy images are collected from a rectangular region of the well. (c) Example images corresponding to experiments initiated with 12,000, 16,000, or 20,000 cells per well. A PC-3 prostate cancer cell line was used. The image recording time is indicated on each subfigure and the scale bar corresponds to 300 μm. The green dashed lines in the images in the top row show the approximate location of the leading edge created by the scratch. Each image is divided into equally-spaced vertical columns, and the number of cells in each column divided by the column area is calculated to yield an estimate of the 1-D cell density.Biophysics, Physiology, Biotechnology, Cancer, Science Policy, Biological Sciences not elsewhere classified, Information Systems not elsewhere classified2020-12-01 18:42:50Generalized Porous-FKPP parameter values.
https://plos.figshare.com/articles/journal_contribution/Generalized_Porous-FKPP_parameter_values_/13315136
Each column corresponds to an experiment with different initial cell density (i.e. 10,000, 12,000, 14,000, 16,000, 18,000, and 20,000 cells per well). (PDF)Biophysics, Physiology, Biotechnology, Cancer, Science Policy, Biological Sciences not elsewhere classified, Information Systems not elsewhere classified2020-12-01 18:42:49Classical FKPP parameter values.
https://plos.figshare.com/articles/journal_contribution/Classical_FKPP_parameter_values_/13315133
Each column corresponds to an experiment with different initial cell density (i.e. 10,000, 12,000, 14,000, 16,000, 18,000, and 20,000 cells per well). (PDF)Biophysics, Physiology, Biotechnology, Cancer, Science Policy, Biological Sciences not elsewhere classified, Information Systems not elsewhere classified2020-12-01 18:42:47BINNs Convergence.
https://plos.figshare.com/articles/figure/BINNs_Convergence_/13315130
Example convergence and improvement plots from training a delay-reaction-diffusion BINN to the scratch assay data with 20,000 initial cells per well. The left subplot shows the training and validation errors (see Eq (3)) in red and blue, respectively, and the black dot shows where the model achieved the best validation error. Similarly, the right subplot shows the training and validation error but only when the error improved. (TIF)Biophysics, Physiology, Biotechnology, Cancer, Science Policy, Biological Sciences not elsewhere classified, Information Systems not elsewhere classified2020-12-01 18:42:46PDE random sampling validation.
https://plos.figshare.com/articles/figure/PDE_random_sampling_validation_/13315127
The BINNs framework is trained using with three ways of including the PDE error term : (a) no PDE regularization, (b) PDE regularization at the data locations, and (c) PDE regularization at 10,000 randomly sampled points at each training iteration. The first column shows the scratch assay data with initial cell density 20,000 cells per well (black dots) with the corresponding BINNs approximation to the governing PDE uMLP (surface plot). The second column shows heatmaps of the modified residual errors (see Eq (6)) at each data point. The third column shows heatmaps of the PDE errors (see Eq (7)) evaluated on a 100 × 100 meshgrid over the input domain. (TIF)Biophysics, Physiology, Biotechnology, Cancer, Science Policy, Biological Sciences not elsewhere classified, Information Systems not elsewhere classified2020-12-01 18:42:44Statistical error model selection.
https://plos.figshare.com/articles/figure/Statistical_error_model_selection_/13315124
The function-approximating deep neural network uMLP is trained using for different values of γ across each data set. Each subplot shows the modified residuals (see Eq (6)) as a function of the predicted cell density u. The columns correspond to different levels of proportionality (i.e. γ = 0.0, 0.2, 0.4, 0.6) where γ = 0.0 represents the constant variance (ordinary least squares) case. Each row (a-f) corresponds to an experiment with different initial cell density (i.e. 10,000, 12,000, 14,000, 16,000, 18,000, and 20,000 cells per well). The proportionality constant that results in the most i.i.d. residuals across each data set was chosen to calibrate the statistical error model in Eq (4). (TIF)Biophysics, Physiology, Biotechnology, Cancer, Science Policy, Biological Sciences not elsewhere classified, Information Systems not elsewhere classified2020-12-01 18:42:43Unconstrained BINN terms.
https://plos.figshare.com/articles/figure/Unconstrained_BINN_terms_/13315121
The learned diffusivity DMLP, growth GMLP, and delay TMLP functions extracted from the corresponding BINNs with governing reaction-diffusion PDE in Eq (9) (first row) and delay-reaction-diffusion PDE in Eq (10) (second row). Each line corresponds to an experiment with a different initial cell density (i.e. 10,000, 12,000, 14,000, 16,000, 18,000, and 20,000 cells per well). Note that DMLP and GMLP have different lengths since they are evaluated between the minimum and maximum observed cell densities corresponding to each data set. (TIF)Biophysics, Physiology, Biotechnology, Cancer, Science Policy, Biological Sciences not elsewhere classified, Information Systems not elsewhere classified2020-12-01 18:42:41Generalized Porous-FKPP model solutions.
https://plos.figshare.com/articles/figure/Generalized_Porous-FKPP_model_solutions_/13315118
Predicted cell density profiles using the Generalized Porous-FKPP model in Eq (14). Each subplot corresponds to an experiment with a different initial cell density (i.e. 10,000, 12,000, 14,000, 16,000, 18,000, and 20,000 cells per well). Solid lines represent the numerical solution to Eq (14) using the parameters that minimize in Eq (6). The markers represent the experimental scratch assay data. (TIF)Biophysics, Physiology, Biotechnology, Cancer, Science Policy, Biological Sciences not elsewhere classified, Information Systems not elsewhere classified2020-12-01 18:42:39Classical FKPP model solutions.
https://plos.figshare.com/articles/figure/Classical_FKPP_model_solutions_/13315115
Predicted cell density profiles using the classical FKPP model in Eq (13). Each subplot corresponds to an experiment with a different initial cell density (i.e. 10,000, 12,000, 14,000, 16,000, 18,000, and 20,000 cells per well). Solid lines represent the numerical solution to Eq (13) using the parameters that minimize in Eq (6). The markers represent the experimental scratch assay data. (TIF)Biophysics, Physiology, Biotechnology, Cancer, Science Policy, Biological Sciences not elsewhere classified, Information Systems not elsewhere classified2020-12-01 18:42:38Spatial errors between BINN solutions.
https://plos.figshare.com/articles/figure/Spatial_errors_between_BINN_solutions_/13315112
Mean GLS errors between the reaction-diffusion and delay-reaction-diffusion BINNs over the spatial dimension for each time point beyond the initial condition. The initial condition is excluded since the PDE solutions are simulated using the initial condition of the data, meaning that the error at t = 0 is zero. Each subplot corresponds to an experiment with a different initial cell density (i.e. 10,000, 12,000, 14,000, 16,000, 18,000, and 20,000 cells per well). (TIF)Biophysics, Physiology, Biotechnology, Cancer, Science Policy, Biological Sciences not elsewhere classified, Information Systems not elsewhere classified2020-12-01 18:42:36Proteomic coverage of the pyruvate dehydrogenase complex.
https://plos.figshare.com/articles/journal_contribution/Proteomic_coverage_of_the_pyruvate_dehydrogenase_complex_/13315114
The proteome was queried for components of the pyruvate dehydrogenase complex and subunits reported with corresponding functions, fold changes, names, and localization to soluble (Sol) or insoluble (Insol). A summary of identified metabolic proteins that relocalized from the soluble to insoluble fraction or vice versa is provided. Fold enrichment significance was generated from DAVID functional annotation clustering and the biological significance indicated with p-value derived from KEGG pathways. Fold changes from non-unique proteins identified in the same metabolic pathway were averaged. (DOCX)Biochemistry, Medicine, Microbiology, Cell Biology, Physiology, Pharmacology, Immunology, Developmental Biology, Cancer, Infectious Diseases, Virology, Computational Biology2020-12-01 18:42:36Infection induced changes in the relative abundance of DC metabolic proteins.
https://plos.figshare.com/articles/journal_contribution/Infection_induced_changes_in_the_relative_abundance_of_DC_metabolic_proteins_/13315108
Influenza virus strain A/PuertoRico/68/34 (IAV) was added to DC for 2 hours, infection medium was replaced, and the infection proceeded for 17 hours followed by cell lysis and protein extraction. Proteins were labeled with SIL or iTRAQ subjected to LC-MS/MS. Both proteomes were combined, redundancies removed, and confidently identified peptides with abundance changes of 2-fold or greater linked to protein identifiers. The lists of upregulated and downregulated proteins were submitted to the Database for Annotation, Visualization, and Integrated Discovery (DAVID) v6.7 and mapped to KEGG pathways. Significantly enriched major metabolic pathways are listed with proteins, soluble (Sol) or insoluble (Insol) fraction and fold change (FC) designated. (DOCX)Biochemistry, Medicine, Microbiology, Cell Biology, Physiology, Pharmacology, Immunology, Developmental Biology, Cancer, Infectious Diseases, Virology, Computational Biology2020-12-01 18:42:34Delay-reaction-diffusion BINN residuals.
https://plos.figshare.com/articles/figure/Delay-reaction-diffusion_BINN_residuals_/13315109
Modified residuals using BINNs with the governing delay-reaction-diffusion PDE in Eq (10). Each subplot corresponds to an experiment with a different initial cell density (i.e. 10,000, 12,000, 14,000, 16,000, 18,000, and 20,000 cells per well). (TIF)Biophysics, Physiology, Biotechnology, Cancer, Science Policy, Biological Sciences not elsewhere classified, Information Systems not elsewhere classified2020-12-01 18:42:34Net changes in DC proteins in response to IAV infection after 17 hours.
https://plos.figshare.com/articles/journal_contribution/Net_changes_in_DC_proteins_in_response_to_IAV_infection_after_17_hours_/13315105
The intensities of the isobaric tag reporter ions were quantified by using the MASIC tool with the exclusion of missing reporter-ion channels or by calculating the SIL ratio for each peptide pair after accounting for singly or doubly labeled species in the 16O/18O ratio and correcting for labeling efficiency. Then, the MS/MS data were searched and filtered by using 0.5% FDR; peptides passing the filter were quantified. Then, peptides-to-protein rollup was performed. (DOCX)Biochemistry, Medicine, Microbiology, Cell Biology, Physiology, Pharmacology, Immunology, Developmental Biology, Cancer, Infectious Diseases, Virology, Computational Biology2020-12-01 18:42:32Delay-reaction-diffusion BINN solutions.
https://plos.figshare.com/articles/figure/Delay-reaction-diffusion_BINN_solutions_/13315106
Predicted cell density profiles using BINNs with the governing delay-reaction-diffusion PDE in Eq (10). Each subplot corresponds to an experiment with a different initial cell density (i.e. 10,000, 12,000, 14,000, 16,000, 18,000, and 20,000 cells per well). Solid lines represent the numerical solution to Eq (10) using TMLP, DMLP, and GMLP. The markers represent the experimental scratch assay data. (TIF)Biophysics, Physiology, Biotechnology, Cancer, Science Policy, Biological Sciences not elsewhere classified, Information Systems not elsewhere classified2020-12-01 18:42:32Reaction-diffusion BINN residuals.
https://plos.figshare.com/articles/figure/Reaction-diffusion_BINN_residuals_/13315103
Modified residuals using BINNs with the governing reaction-diffusion PDE in Eq (9). Each subplot corresponds to an experiment with a different initial cell density (i.e. 10,000, 12,000, 14,000, 16,000, 18,000, and 20,000 cells per well). (TIF)Biophysics, Physiology, Biotechnology, Cancer, Science Policy, Biological Sciences not elsewhere classified, Information Systems not elsewhere classified2020-12-01 18:42:31Reaction-diffusion BINN solutions.
https://plos.figshare.com/articles/figure/Reaction-diffusion_BINN_solutions_/13315100
Predicted cell density profiles using BINNs with the governing reaction-diffusion PDE in Eq (9). Each subplot corresponds to an experiment with a different initial cell density (i.e. 10,000, 12,000, 14,000, 16,000, 18,000, and 20,000 cells per well). Solid lines represent the numerical solution to Eq (9) using DMLP and GMLP. The markers represent the experimental scratch assay data. (TIF)Biophysics, Physiology, Biotechnology, Cancer, Science Policy, Biological Sciences not elsewhere classified, Information Systems not elsewhere classified2020-12-01 18:42:29Simulation parameter fits.
https://plos.figshare.com/articles/figure/Simulation_parameter_fits_/13315097
The learned diffusivity and growth functions DMLP and GMLP evaluated over cell density u. Starting from the left, the first two subplots correspond to the learned diffusivity and growth functions from simulated data using the classical FKPP equation. The last two subplots correspond to the learned diffusivity and growth functions from simulated data using the Generalized Porous-FKPP equation. Solid lines represent the parameter networks DMLP and GMLP and dashed lines represent the true diffusivity and growth functions used to simulate the data. (TIF)Biophysics, Physiology, Biotechnology, Cancer, Science Policy, Biological Sciences not elsewhere classified, Information Systems not elsewhere classified2020-12-01 18:42:28Simulation model fits.
https://plos.figshare.com/articles/figure/Simulation_model_fits_/13315094
Predicted cell density profiles using BINNs with the governing reaction-diffusion PDE in Eq (9). The left subplot corresponds to the set of simulated data using the classical FKPP equation and the right subplot corresponds to the Generalized Porous-FKPP equation. Solid lines represent the numerical solution to Eq (9) using DMLP, and GMLP. Dashed lines represent the noiseless numerical simulations of the classical FKPP and Generalized Porous-FKPP equations. The markers represent the numerical simulations of the classical FKPP and Generalized Porous-FKPP equations with artificial noise generated by the statistical error model in Eq (4). (TIF)Biophysics, Physiology, Biotechnology, Cancer, Science Policy, Biological Sciences not elsewhere classified, Information Systems not elsewhere classified2020-12-01 18:42:26Neural diffusivity and pre-emptive epileptic seizure intervention
https://plos.figshare.com/collections/Neural_diffusivity_and_pre-emptive_epileptic_seizure_intervention/5225180
The propagation of epileptic seizure activity in the brain is a widespread pathophysiology that, in principle, should yield to intervention techniques guided by mathematical models of neuronal ensemble dynamics. During a seizure, neural activity will deviate from its current dynamical regime to one in which there are significant signal fluctuations. In silico treatments of neural activity are an important tool for the understanding of how the healthy brain can maintain stability, as well as of how pathology can lead to seizures. The hope is that, contained within the mathematical foundations of such treatments, there lie potential strategies for mitigating instabilities, e.g. via external stimulation. Here, we demonstrate that the dynamic causal modelling neuronal state equation generalises to a Fokker-Planck formalism if one extends the framework to model the ways in which activity propagates along the structural connections of neural systems. Using the Jacobian of this generalised state equation, we show that an initially unstable system can be rendered stable via a reduction in diffusivity–i.e., by lowering the rate at which neuronal fluctuations disperse to neighbouring regions. We show, for neural systems prone to epileptic seizures, that such a reduction in diffusivity can be achieved via external stimulation. Specifically, we show that this stimulation should be applied in such a way as to temporarily mirror the activity profile of a pathological region in its functionally connected areas. This counter-intuitive method is intended to be used pre-emptively–i.e., in order to mitigate the effects of the seizure, or ideally even prevent it from occurring in the first place. We offer proof of principle using simulations based on functional neuroimaging data collected from patients with idiopathic generalised epilepsy, in which we successfully suppress pathological activity in a distinct sub-network prior to seizure onset. Our hope is that this technique can form the basis for future real-time monitoring and intervention devices that are capable of treating epilepsy in a non-invasive manner.Cell Biology, Neuroscience, Physiology, Cancer, Biological Sciences not elsewhere classified, Mathematical Sciences not elsewhere classified, Information Systems not elsewhere classified2020-12-01 18:42:13Ranking order in terms of the stimulation strength required to suppress the BOLD response in the mid-frontal region to the same extent as in Fig 8B, by stimulating a single region in the seizure network.
https://plos.figshare.com/articles/dataset/Ranking_order_in_terms_of_the_stimulation_strength_required_to_suppress_the_BOLD_response_in_the_mid-frontal_region_to_the_same_extent_as_in_Fig_8B_by_stimulating_a_single_region_in_the_seizure_network_/13315091
Stimulation strengths are presented relative to the lowest value (thalamus left), which is assigned a value of unity. N/A values are assigned if it is not possible to suppress the BOLD response in the mid-frontal region by targeting the corresponding single region.Cell Biology, Neuroscience, Physiology, Cancer, Biological Sciences not elsewhere classified, Mathematical Sciences not elsewhere classified, Information Systems not elsewhere classified2020-12-01 18:42:13Simulating seizure suppression.
https://plos.figshare.com/articles/figure/Simulating_seizure_suppression_/13315088
A) External stimulation signal intensity (i), normalized between zero and unity, applied to all nodes except for the mid frontal region for 1000 forward models ranging from zero stimulation (yellow) to a stimulation profile that perfectly mirrors the pre-ictal activity in Fig 7B (black). B) Response signal intensity (i), normalized between zero and unity, of the mid frontal region for the same 1000 forward models in A) with matching colours, i.e. the yellow response corresponds to zero stimulation and the black response corresponds to a stimulation profile that perfectly mirrors the pre-ictal activity in Fig 7B.Cell Biology, Neuroscience, Physiology, Cancer, Biological Sciences not elsewhere classified, Mathematical Sciences not elsewhere classified, Information Systems not elsewhere classified2020-12-01 18:42:11Fate of antibiotic resistant E. coli and antibiotic resistance genes during full scale conventional and advanced anaerobic digestion of sewage sludge
https://plos.figshare.com/collections/Fate_of_antibiotic_resistant_i_E_i_i_coli_i_and_antibiotic_resistance_genes_during_full_scale_conventional_and_advanced_anaerobic_digestion_of_sewage_sludge/5225102
Antibiotic resistant bacteria (ARB) and their genes (ARGs) have become recognised as significant emerging environmental pollutants. ARB and ARGs in sewage sludge can be transmitted back to humans via the food chain when sludge is recycled to agricultural land, making sludge treatment key to control the release of ARB and ARGs to the environment. This study investigated the fate of antibiotic resistant Escherichia coli and a large set of antibiotic resistance genes (ARGs) during full scale anaerobic digestion (AD) of sewage sludge at two U.K. wastewater treatment plants and evaluated the impact of thermal hydrolysis (TH) pre-treatment on their abundance and diversity. Absolute abundance of 13 ARGs and the Class I integron gene intI1 was calculated using single gene quantitative (q) PCR. High through-put qPCR analysis was also used to determine the relative abundance of 370 ARGs and mobile genetic elements (MGEs). Results revealed that TH reduced the absolute abundance of all ARGs tested and intI1 by 10–12,000 fold. After subsequent AD, a rebound effect was seen in many ARGs. The fate of ARGs during AD without pre-treatment was variable. Relative abundance of most ARGs and MGEs decreased or fluctuated, with the exception of macrolide resistance genes, which were enriched at both plants, and tetracyline and glycopeptide resistance genes which were enriched in the plant employing TH. Diversity of ARGs and MGEs decreased in both plants during sludge treatment. Principal coordinates analysis revealed that ARGs are clearly distinguished according to treatment step, whereas MGEs in digested sludge cluster according to site. This study provides a comprehensive within-digestor analysis of the fate of ARGs, MGEs and antibiotic resistant E. coli and highlights the effectiveness of AD, particularly when TH is used as a pre-treatment, at reducing the abundance of most ARGs and MGEs in sludgeand preventing their release into the environment.Biochemistry, Microbiology, Ecology, Infectious Diseases, Plant Biology, Virology, Environmental Sciences not elsewhere classified, Biological Sciences not elsewhere classified, Chemical Sciences not elsewhere classified2020-12-01 18:42:10Pre-ictal perturbation.
https://plos.figshare.com/articles/figure/Pre-ictal_perturbation_/13315085
A) EEG activity from the frontal lobe (FP1 & FP2). Ictal activity is shown by the yellow section and pre-ictal activity is shown by the red section. B) The seizure network shown in MNI space (left) with the mid-frontal regions (left & right) indicated by the red nodes. The mean pre-ictal signal intensity (i) (right), normalized between zero and unity, corresponding to the red sections in A), is used as the external driving input and is supplied to the red nodes, as indicated by the inward-pointing red arrows. C) The mean BOLD signal intensity (i), normalized between zero and unity, of the mid-frontal region to the stimulus in B), as indicated by the outward-pointing red arrows.Cell Biology, Neuroscience, Physiology, Cancer, Biological Sciences not elsewhere classified, Mathematical Sciences not elsewhere classified, Information Systems not elsewhere classified2020-12-01 18:42:10Fig 7 -
https://plos.figshare.com/articles/figure/Fig_7_-/13315084
A. PCoA of ARGs within sludge samples. B. PCoA of MGEs within sludge samples. Each colour shows an independent sampling event. Samples taken from the same point in the treatment process are indicated using the same shapes.Biochemistry, Microbiology, Ecology, Infectious Diseases, Plant Biology, Virology, Environmental Sciences not elsewhere classified, Biological Sciences not elsewhere classified, Chemical Sciences not elsewhere classified2020-12-01 18:42:09Bayesian model reduction.
https://plos.figshare.com/articles/figure/Bayesian_model_reduction_/13315082
A) Control group. Approximate lower bound on log model evidence afforded by the free energy (F) following Bayesian model reduction for the reduced model without (w/o) diffusion and the full model with (w) diffusion. B) Probabilities derived from the log evidence in A). C) & D) Same layout as A) & B), but for the patient group.Cell Biology, Neuroscience, Physiology, Cancer, Biological Sciences not elsewhere classified, Mathematical Sciences not elsewhere classified, Information Systems not elsewhere classified2020-12-01 18:42:08Diversity of ARGs (A) and MGEs (B) during sludge treatment. Samples 1–9 are from WWTP 1.
https://plos.figshare.com/articles/figure/Diversity_of_ARGs_A_and_MGEs_B_during_sludge_treatment_Samples_1_9_are_from_WWTP_1_/13314374
Samples 1, 4, 7 are influent sludge, samples 2, 5, 8 are post-TH sludge, samples 3,6,9 are post-MAD sludge. Each sample set (1–3, 4–6, 7–9) is from an independent sampling event. Samples 10–15 are from WWTP 2. Samples 10, 12, 14 are influent sludge, samples 11,13,15 are post-MAD sludge. Each sample set (10 and 11, 12 and 13, 14 and 15) is from an independent sampling event.Biochemistry, Microbiology, Ecology, Infectious Diseases, Plant Biology, Virology, Environmental Sciences not elsewhere classified, Biological Sciences not elsewhere classified, Chemical Sciences not elsewhere classified2020-12-01 18:42:07Stability.
https://plos.figshare.com/articles/figure/Stability_/13315079
The sum of the Real components of the eigenvalues of the Jacobian – measuring intrinsic stability of neural dynamics–as a function of structural (DTI) adjacency matrix threshold (%) for patients and controls following Bayesian model averaging. The 90% Bayesian credible intervals are sufficiently small to be contained within the data points.Cell Biology, Neuroscience, Physiology, Cancer, Biological Sciences not elsewhere classified, Mathematical Sciences not elsewhere classified, Information Systems not elsewhere classified2020-12-01 18:42:06Diffusion coefficients.
https://plos.figshare.com/articles/figure/Diffusion_coefficients_/13315076
The diffusion coefficient (σ) (following Bayesian model averaging) as a function of structural (DTI) adjacency matrix threshold (%) for patients and controls. The 90% Bayesian credible intervals are sufficiently small to be contained within the data points.Cell Biology, Neuroscience, Physiology, Cancer, Biological Sciences not elsewhere classified, Mathematical Sciences not elsewhere classified, Information Systems not elsewhere classified2020-12-01 18:42:04