Journal: Neuroinformatics

Article Title: A Modular Workflow for Model Building, Analysis, and Parameter Estimation in Systems Biology and Neuroscience

doi: 10.1007/s12021-021-09546-3

Figure Lengend Snippet: An overview of available software packages and tools used by the Subcellular Workflow. Sometimes, manual intervention is needed in between workflow modules. This is especially true when an input format does not have a feature that an output format has. Additionally, some software packages have so many functions that it could be easier to use them interactively (in these cases we also added a yes in the manual intervention column)

Article Snippet: , MATLAB® Optimization ToolboxTM , MATLAB® , Parameter estimation , none , none , no.

Techniques: Software, Plasmid Preparation, Functional Assay

Journal: Neuroinformatics

Article Title: A Modular Workflow for Model Building, Analysis, and Parameter Estimation in Systems Biology and Neuroscience

doi: 10.1007/s12021-021-09546-3

Figure Lengend Snippet: A Illustration of the model inputs. Calcium burst (blue) at 4 s used in all simulations and a dopamine transient (orange) applied at different timings in eight experiments and one without it. B Four species used in parameter estimation corresponding to the input combination in A. Black traces represent the data produced by simulating the original model, red traces represent fits with the best new parameter sets in the updated model. C , D Comparison of model performance with substrate phosphorylation as the main model readout. Here, 30 s simulations were used for comparison with the original model behavior. C Normalized time series of substrate phosphorylation, the main readout, with calcium as an only input or dopamine following it after 1s. D Normalized area under the curve of substrate phosphorylation with different calcium and dopamine input intervals. The simulations performed to obtain these graphs in MATLAB® took less than 5 min to compute (intel core i9-10980XE)

Article Snippet: , MATLAB® Optimization ToolboxTM , MATLAB® , Parameter estimation , none , none , no.

Techniques: Produced, Comparison, Phospho-proteomics

Journal: Neuroinformatics

Article Title: A Modular Workflow for Model Building, Analysis, and Parameter Estimation in Systems Biology and Neuroscience

doi: 10.1007/s12021-021-09546-3

Figure Lengend Snippet: Simulations in identical conditions in both MATLAB® SimBiology® and COPASI yielded almost identical results. A Inputs used in both simulators. The calcium input is kept constant at 4 s for all simulations and dopamine input time is varied from time 0 to 8 s at every one second. The difference from the previous simulations is in the calcium input which, for the sake of simplicity, is represented by a double exponential spike. B and C show substrate phosphorylation curves analogous to Fig. , the red line represents results obtained in MATLAB® and blue line results from simulations in COPASI. A single 30 s simulation took around 10 s of compute time within a 1 fl spine volume (Intel® Core™ i7-8750H)

Article Snippet: , MATLAB® Optimization ToolboxTM , MATLAB® , Parameter estimation , none , none , no.

Techniques: Phospho-proteomics

Journal: Neuroinformatics

Article Title: A Modular Workflow for Model Building, Analysis, and Parameter Estimation in Systems Biology and Neuroscience

doi: 10.1007/s12021-021-09546-3

Figure Lengend Snippet: Validation of the model by stochastic STEPS simulation of substrate phosphorylation in a typical D1 MSN spine. A Normalized time course of substrate phosphorylation in the updated model run in MATLAB® in comparison with averaged (n = 50) stochastic STEPS simulations (red–calcium and dopamine; blue–calcium only) for a typical size of D1 MSN synaptic spine (V = 0.02μm3). The same stimulation protocol as in Fig. was used. Colored areas around averaged STEPS curves correspond to a range between 10 and 90% confidence intervals. One simulation required less than 1 min of compute time for 30 s of simulated reactions within a 0.02 µl spine (~6000 molecules). B Normalized area under the curve of substrate phosphorylation with different calcium and dopamine input intervals simulated for the MATLAB® version of the updated model (with MATLAB® ode15s solver, black line) and averaged stochastic STEPS simulations (n = 30) in the application version of the model. MATLAB® statistical bar plots were added to the figure to characterize variability of synaptic plasticity between subsequent induction protocol applications to the same synaptic spine. Note that despite high variability of synaptic plasticity time courses averaged plasticity dynamics were in a good agreement with the ODE-based solution

Article Snippet: , MATLAB® Optimization ToolboxTM , MATLAB® , Parameter estimation , none , none , no.

Techniques: Biomarker Discovery, Phospho-proteomics, Comparison

Journal: Neuroinformatics

Article Title: A Modular Workflow for Model Building, Analysis, and Parameter Estimation in Systems Biology and Neuroscience

doi: 10.1007/s12021-021-09546-3

Figure Lengend Snippet: Inserting the biochemical signal transduction cascade into an electrical model in NEURON. A A schematic of the effects of the two inputs of the model, dopamine and calcium, on a generic substrate, which in this case is taken to represent the fraction of phosphorylated AMPA receptors with higher conductance levels. B Examples of the two inputs, calcium and dopamine. The calcium signal at the synapse is a result of ten repeats of a synaptic stimulus paired with three somatic spikes evoked with a current clamp (Yagishita et al., , Fig. ). C The simulations in MATLAB and NEURON give the same results when using the same calcium input from the NEURON simulation in MATLAB. This simulation in NEURON requires 4 h on 8 compute nodes on an Intel i7-4700MQ CPU @ 2.40 GHz. and less than one second in MATLAB® using an intel core i9-10980XE D , E Predicted EPSP (Excitatory PostSynaptic Potential) following a single synaptic input in the relevant spine and in the soma. The readout of the substrate phosphorylation level was done at 7 s after the start of the dopamine input. The relative timings of dopamine and calcium indicated in the figure legends are used, and the results are compared to the experimental setting without the dopamine input

Article Snippet: , MATLAB® Optimization ToolboxTM , MATLAB® , Parameter estimation , none , none , no.

Techniques: Transduction, Phospho-proteomics