Physiology and Morphology
UNDER CONSTRUCTION BY AARONS - PLEASE DO NOT EDIT
Searching the Database
In the current release of the Allen Cell Types Database, we include electrophysiological recordings from 1058 mouse cells and 279 human cells. For a subset of these cells, we also include morphological reconstructions and neuronal models. The Cell Feature Search page allows you to select cells that satisfy certain conditions and to view summary cards for those cells in a desired sorting order. From a cell's summary card, you can then navigate to pages that contain detailed electrophysiological or morphological information for that cell.
The Cell Feature Search page is divided into three areas: 1) In the top area, you can set filters for cell properties that have discrete values; for example, you can choose to select only cells from a certain species and certain cortical layers. You can also use this page to find cells for which specific types of data are available; for example, you can select only cells for which there is not only electrophysiological data, but also morphological reconstructions and neuronal models. 2) In the middle area, you can set conditions on numerical features in certain value ranges by using a parallel coordinate plot. 3) In the bottom area, you can view the summary cards for the cells that were selected with the filters set in the two areas above and you can then choose the sorting order in which those cards are displayed.
You can set filters for cell properties or data types by checking boxes and selecting one or more items from the drop-down menus at the top of the Cell Feature Search page. (Note: while the Transgenic Targeting and Donor Profile selections apply only to mouse and human cells, respectively, the rest apply to all cells.)
The following table describes the meaning of the properties available for filtering:
|Property name||Property description|
|Species||Species of recorded cell|
|Area||Brain region of recorded cell|
|Layer||Cortical layer of recorded cell|
|Hemisphere||Hemisphere of recorded cell|
|Mouse line||Transgenic driver of reporter expression|
|Reporter status||Expression of transgenic reporter|
|Age||Age of human donor|
|Years of seizure||Length of seizure history|
|Donor||Human donor ID|
|Sex||Sex of human donor|
|Disease||Disease state of human donor|
|Ethnicity/Race||Ethnicity/race of human donor|
|Reconstruction type||Extent of morphological reconstruction|
|Dendrite type||Status of dendritic spines|
|Apical dendrite status||Extent of apical dendrite preservation, if applicable|
|Has GLIF model||Existence of GLIF model for cell|
|Has all-active biophysical model||Existence of all-active biophysical model for cell|
|Has perisomatic biophysical model||Existence of perisomatic biophysical model for cell|
Parallel Coordinate Plot Of Cell Features
You can use the parallel coordinate plot to select cells with numerical features that satisfy certain conditions. First you choose up to five different fetures from the drop down menus under the axes, and then you select the value ranges of interest by clicking on the axes. For example, the cells selected in the figure below have Upstroke:Downstroke ratios in the range 2.0 to 5.0 and Adaptation indeces in the range -0.4 to 0.4. Each line represents a cell and the intersection with each axis indicates the value of the feature associated with that axis. Cells that do not have feature values in the selected ranges are represented by gray lines.
The "color by" drop-down menu allows you to select a feature whose value determines the line color for the cell. For example, in the figure below the lines are colored according to the Upstroke:Downstroke ratio values, so cells with higher values are represented by purple lines while cells with lower values are represented by blue lines.
The following sections describe the cell features that can be selected with the Parallel Coordinate Plot.
Cell Feature Filters
While many distinct electrophysiological features were recorded and/or calculated from each cell, only a few features were called out in the web application. To see the other features, please read the Electrophysiology Overview whitepaper in the "Documentation" tab and/or download the data from the Allen Brain Atlas API. These features were selected not only by what they were able to tell us about the intrinsic properties of the cells, but also in their ability to differentiate between putative cell types in a non-biased clustering analysis.
The ratio between the absolute values of the action potential peak upstroke and the action potential peak downstroke. Action potential peak upstroke: The maximum rate of change between the action potential threshold and the action potential peak. Action potential peak downstroke: The minimum rate of change between the action potential peak and the action potential trough.
This value gives us an indication of the amount of time a cell takes to recover after an Action Potential. An example of the difference between a high and a low ratio is illustrated here.
This parameter is used to distinguish between spiny (putatively excitatory) and aspiny (putatively inhibitory) groups of cells in a non-biased clustering analysis.
Not only are isolated action potentials of interest, but the rates of firing are likely to be important in inter-neuron communication. The adaptation index is the rate at which action potential firing changes during a stimulus. This index is used to characterize a neuron's response to a sustained input, a one second square pulse in this case.
A high adaptation index means the neuron spikes at the beginning of the stimulus, but then the spiking slows or stops altogether. These neurons therefore likely transmit the initiation of an upstream input. Conversely, a neuron that does not accommodate (has a low adaptation index) gives a consistent output during the sustained input and might also be used to tell when a stimulus ends. The adaptation index likely involves ion channel proteins that change properties on the time scale of hundreds of milliseconds to seconds.
This value is the amount of stimulus current (long square current in pA) required to initiate a single action potential. This parameter gives us an indication of the excitability and the membrane resistance of the cell.
Membrane Time Constant
input resistance and the cell capacitance. The membrane time constant in this study was estimated by fitting voltage responses to negative current injection with an exponential.The membrane time constant, also referred to as tau, reflects the time it takes for a neuron to charge its membrane. This is one of the passive or subthreshold properties of the cell and is the product of the
parallel coordinate plot where cells with a greater Upstroke/Downstroke ratio show a lower firing rate.The firing rate is simply the speed at which a neuron fires multiple action potentials in sequence given a sustained current injection. A neuron that fires at a faster rate could influence a downstream neuron more strongly than a neuron firing with a slower rate. The rate that a neuron fires given a sustained input is determined by the ion channel proteins it expresses and whether and how fast those ion channel proteins recover from the previous action potential. In general, a higher firing rate is seen in neurons with individual action potentials that are more narrow and have a deeper after hyperpolarization. This is illustrated in the
tau is one of the passive or subthreshold properties of the cell. A cell with high input resistance leads to a large voltage response (V=IR) and is referred to as "tight" as current doesn't easily leak out. Conversely, "leaky" neurons are those with low input resistance. The input resistance was calculated by measuring the slope of the V/I plot at subthreshold potentials.The input resistance is the baseline resistance of the neuron membrane, and like
A subset of cells with electrophysiological recordings was manually selected for 3-D reconstruction. Similar to the electrophysiology features, many morphology features were calculated in the reconstruction, but only three are called out in the web application; Normalized Cortical Depth, Max Distance and # Stems. To see the entire array of morphological features, please read the Morphology Overview whitepaper in "Documentation" or download the data from the Allen Brain Atlas API.
The three features represented in the web application were the top three most discriminative features (ranked by a classic machine learning technique) that best separates major morphological categories of neurons currently included in the database.
Normalized Cortical Depth
As the thickness of the visual cortex changes naturally and can also be affected by non-biological factors like tissue shrinkage during histology, we compute the relative depth of the soma with respect to the pia and white matter using the cortical coordinate space in the CCF (see whitepaper in Documentation).
This is the maximum Euclidean distance of all the nodes. Euclidean distance is the straight line distance from the soma (root) to the node. The maximum Euclidean distance is able to distinguish a particular category of spiny neuron ("tall, tufted") from a particular category of aspiny neuron ("common type").
Number of Stems
The number of stems attached to the soma. The # Stems is used to distinguish between two categories of spiny neurons (“star pyramid” versus “tufted”).
Filtering of the dataset results in a list of experiments ranked (by default) by Upstroke:Downstroke. You can alter the sort parameters from the Filters menu.
features are calculated and (when available) the reconstructed neuron.Each experiment lists metadata regarding the classification of the cell, including the Cre-line, the Area, the Layer, Cell Reporter (positive or negative), the Dendrite Type as well as whether or not the Apical Dendrite is intact. It also includes thumbnails showing a trace of the action potential from which many of the electrophysiological
The thumbnail depicting the reconstructed neuron includes the dendrites (red), the apical dendrites (orange) and the axon (blue). The portion of the neuron representing each of these morphologies is depicted in a histogram to the right of the reconstructed neuron. The neurons location in the cortex is indicated by the scale showing normalized cortical depth where the top is the pial surface and the bottom is the white matter. NOTE: this scale is not a measure of cortical layer.
If you have selected several cells to compare, you can save this or other kinds of searches by selecting the "Permalink" feature.
Clicking anywhere in a cell's results box will open a new page with the experimental details.
Experimental Detail Page - Electrophysiology
The Experimental Detail page includes the Electrophysiological Summary and a workspace to Browse Electrophysiology Data. The Electrophysiology Summary includes a thumbnail of the location of the cell mapped to the CCF, metadata on the cell including Mouse Line, ID, Area, Cell Reporter, Dendrite Type, Apical Dendrite and Hemisphere. Values from each of the electrophysiology cell features are also listed along with model parameters (where appropriate). You can also view plots of the F/I and V/I curves here.
Clicking on the image of the cell (when available) will open a new page containing the Morphology data.
The electrophysiological data itself can be viewed from this page, or downloaded to be visualized on another platform or in a third party program. The Allen SDK provides a simple Python module to support downloading metadata and NWB files for cells in the Allen Cell Types Database. Please see the Data API Client documentation page to see an example.
- Select Stimulus type: A drop-down menu from which you can select the stimulus type and see the resulting Cell Response.
- Select Neuronal Model: When available, neuronal models have been run on the data and when selected will open below the recorded Cell Response.
- Download Data: This link will download the .nwb file with the data from this experiment. For more information, please see here.
- Select Sweep: Select sweeps will be available for you to inspect from this view. As you hover your mouse over each colored square, not only do you see the resulting graphs change to reflect the sweep selected, but you also will see sweep metadata (Sweep #, Stimulus amplitide (pA) and # of spikes) listed below the squares. Once you click on a colored square, you can use left/right arrow keys to move between the sweeps.
- Slider Bar: This feature allows you to zoom in and out of the Stimulus, Cell Response and Model views by clicking and dragging on the arrows.
- Stimulus: The stimulus injected into the cell.
- Cell Response: The response of the cell to the injected stimulus.
- (Optional) Model Response: When available, the model of the cellular response.
Different sets of stimulation waveforms were used in order to:
- Interrogate intrinsic membrane mechanisms that underlie the input/output function of neurons
- Linear and non-linear subthreshold properties
- Action potential initiation and propagation
- Understand aspects of neural response properties in vivo
- Stimulation frequency dependence (theta vs. gamma) of spike initiation mechanisms
- Ion channel states due to different resting potentials in vivo
- Construct and test computational models of varying complexity emulating the neural response to stereotyped stimuli
- Generalized leaky-integrate-and-fire (GLIF) models
- Biophysically and morphologically realistic conductance-based compartmental models
Reprocessing of the data occurred for the March 2016 release so any analysis performed prior to the March 2016 release date should be performed again with the new models.
The Allen Cell Types Database contains three types of neuronal models: two biophysical models and generalized leaky integrate-and-fire (GLIF) models. These models attempt to mathematically reproduce a cell's recorded response to a current injection. The biophysical models take into account dendritic morphological structure, whereas GLIF models are simple point neuron models that represent the neuron as a single compartment.
There are five levels of GLIF models with increasing levels of complexity. The most basic model is a simple leaky integrate-and-fire equation. More advanced GLIFs attempt to model variable spike threshold, afterspike currents, and threshold adaptation.
More detailed information on each of the models is available in the whitepapers in Documentation.
1. Leaky Integrate and Fire (LIF)
Standard circuit representation of a resistor and capacitor in parallel with a leaky membrane.
2. LIF + Reset Rules (LIF-R)
LIF with biologically-derived threshold and voltage reset rules in addition to a biologically derived threshold decay.
3. LIF + Afterspike Currents (LIF-ASC)
LIF with spike-induced currents to model long-term effects of voltage-activated ion channels.
4. LIF-R + Afterspike Currents (LIF-R-ASC)
LIF with additional Reset Rules and Afterspike Currents.
5. LIF-R-ASC + Threshold Adaptation (LIF-R-ASC-A)
All of the above, with an additional voltage-dependent component of threshold.
Biophysically realistic, single-neuron model with passive dendrites and active soma.
Biophysically realistic, single-neuron model with active conductances everywhere.
When available: the Morphology page can be reached either by clicking on the image of the reconstructed cell in the search results page,
or by clicking on the image of the cell on the Electrophysiological Detail Page.
Experimental Detail Page - Morphology
The Experimental Detail page includes the morphology summary and a viewer to browse the morphology data. The Morphology Summary includes the location of the cell mapped to the CCF, metadata on the Mouse Line and location of the cell. It also shows the Morphological features most appropriate to the data (see more information in the Morphology whitepaper in Documentation), a thumbnail showing the neuron reconstruction including cortical depth, as well as a thumbnail illustrating the electrophysiology traces.
Clicking on the electrophysiology thumbnail will open a new page containing the Electrophysiology Results data.
Projected views of the biocytin filled neuron were constructed by composing the darkest intensity pixel from each plane of the image stack into a single plane.
Projected top view and side view, as well as the 3-D neuron reconstruction can be viewed from this page.
From the Projected top view, you can zoom into the picture from the on-screen navigation tools, the Keyboard Commands or using your scroll wheel. The two views of the neuron are synched so zooming in on one will also zoom the other. Clicking on "View image stack" will take you to an image viewer to view the individual images taken of this neuron.
The image viewer of the 3-D neuron reconstruction allows for visualization of the reconstructed neuron using the onscreen navigation tools. Clicking "Reset" will reset the neuron to its default view. The legend in the 3D reconstruction indicates the various components of the reconstruction.
You can download both the reconstruction (as an .swc file) or the calculated morphological measurements (as an XML) from the links below the viewers. For more information please see the API Documentation.
Morphology Image Stack
Clicking "View Image Stack" while browsing the Morphology data will take you to our image viewer. The title bar includes the Mouse Line, the Specimen ID, the structure and the hemisphere. The "Configure" icon opens a menu that will allow you to vary the image contrast and download the individual images. The entire image stack can be navigated through using the on screen navigation tools, using the Keyboard Commands or by clicking on the Projected Side View.
Shows the current viewing resolution of the image, in microns. This value dynamically changes as you zoom in/out of the image. You can position the scale bar anywhere on the main image by dragging the scale bar by its ruler.
You can toggle the orientation of the scale bar from horizontal to vertical by clicking on the scale bar text.
Advance to the next image from the specimen
Go back to the previous image from the specimen