This schema specifies basic common elements that most BrainML schemas are
expected to use and inherit from.
Nothing is specific to a particular field of neuroscience. Included are:
View and Trace definitions
Basic recording site definitions
Basic protocol definition
Experiment definition
Link that must go to an experiment entity.
Link that must go to a trace entity.
Link going to a trace that serves as stimulus for containing trace, which
should be a response.
Link going to a trace that serves as response for containing trace, which
should be a stimulus.
Link that must go to a recording_site entity.
Contains a set of links to traces forming a logical group.
A short description of the grouping type, e.g. 'analysis' or
'simultaneous' etc..
Short name/description of a View or Trace.
Unit of measure for horizontal axis (often a temporal unit).
Short, informative label for horizontal axis.
Unit of measure for vertical axis (dependent variable).
Short, informative label for vertical axis.
Recording start time for Trace.
Recording end time for Trace.
Indication of whether Trace represents stimulus signal or neuron response.
Trace sampling rate, in units of horizontal axis.
Number of trials contributing to View or Histogram's data.
Type of units reflected by bar height: 'probability', 'percentage', 'Hz',
or 'count'.
First bin center for Histogram (in H-axis units).
Width of bin in Histogram (in H-axis units).
Number of bins in Histogram.
Whether or not the first and last bar heights reflect measurement including
h-axis values less than min or greater than max.
Minimum h-axis value in raw histogram.
maximum h-axis value in a raw histogram.
Paragraph-length description of Experiment.
Recording technique employed in collecting Trace.
The nature of the data collected in Trace.
Ordering position of this View within its containing experiment.
Superclass View element. Do not use directly, but any View defined should
declare itself in a substitution group with this.
Not for inheritance, since we can't later say that the slot must be a
subclass so that only certain trace types can appear within certain views.
View for containing time-series data.
View for containing X-Y (2-dimensional plot) data. Traces share common
X- and Y-axis units.
View for containing histogram data.
Superclass Trace element. Do not use directly.
Order of this Trace relative to others in the same View.
Superclass with common elements of all traces valid in time series view.
Trace consisting of a regularly-sampled list of data values.
Trace consisting of a list of event times at which spikes occurred.
Trace consisting of a list of event times and describing labels.
These labels may be either strings, integers, or decimal values.
Data should be contained in a 'labeled_dataset' element, whose first
child is a decimal type dataset, and whose second child is a dataset
of any type.
The piecewise_series trace is a time series in which flat or linear sections
are encoded compactly. The data consists of a list of segments. Each segment
starts with a type code and a duration. Type code may be 1, indicating a
constant-value segment, 2, indicating a linear segment, 3, indicating a
normal (fully-specified) segment, or 4, indicating an empty segment (gap).
The duration is an integer count indicating the number of samples specified
by this segment. For constant-value segments, the next number is the
constant value to be held for this segment. For linear segments, the next
number is the final value this segment reaches at the end of its duration.
The values before it are to be interpolated linearly starting from the value
of the series just before the linear segment begins.
(A linear segment cannot begin a piecewise series -- use a length-0
constant-value segment to indicate the starting value if needed.)
For type 3 segments, the next duration numbers are the actual samples for
this segment.
For type 4 segments, there are no numbers following the duration. Instead,
the time series is understood to have no values (not been recorded) for the
given duration.
Bivariate trace consisting of a list of x- and y-values. Dataset 'dimensions'
should have two values, the first indicating the number of data points in the
trace (or *), the second indicating the size of each tuple. This should be
2 for plain (x,y) pairs, 3 for (x, y, y-error), 4 for
(x, y, y-error, x-error). To use asymmetrical error bars, set second
dimension to 6 and use the format
(x, y, +y-error, -y-error, +x-error, -x-error).
Superclass for histogram traces.
Prebinned histogram trace. Dataset 'dimensions' should have two values, the
first indicating the number of data points in the trace (or *), the second
indicating the size of each tuple, either 1 (value), 2 (value, +/- error),
or 3 (value, + error, - error).
Raw histogram trace that the end-consumer selects bin size for. The values
provided here are raw values (e.g., the list of times after a marker that
spikes occurred). The binning process counts up the numbers of values
within each range.
Holds an entire submission.
Account name (for submitters of data submissions).
A person who signed in and submitted data.
A person contributing to a data submission.
Contains information on recording site (location and source).
Short, unique, human-readable name or code that should be same as any
references in published work.
Description of the various recording locations.
Keywords describing brain or body location(s).
Cytoarchitectonic area recording site was found in (Brodmann scheme).
Description of the recording layers.
Description of the cell types.
Superclass for recording sources; do not use directly.
Base entity for describing the subject or preparation in an experiment.
Different models will extend this depending on what descriptors they
require.
Experimental protocol.
Preparation type used in this Protocol.
Short verbal description of Protocol.
Holds details on subject preparation and response elicitation.
The agent effecting one aspect of the Protocol.
The manner in which an agent acts to effect one aspect of the Protocol.
The locus at which an agent acts to effect one aspect of the Protocol.
Experimental condition. This is a generic structure for holding information
about the conditions under which data is collected that is not otherwise
represented in the data model. This provides a dimension of flexibility to
allow recording of information not anticipated at the time of data model
construction, however it is generally best to avoid use of this facility when
possible because it prevents standardization on descriptors used. This makes
data more difficult to search for and link to data in other databases.
A short descriptive name of what this condition represents.
Data type for this condition's value: either 'numeric' or 'text'.
Value for the experimental condition. Should follow 'type'.
For numeric condition values that represent measured quantities, specify
the units. Note this is not controlled in the way units generally are in
BrainML (based on the BrainMetaL 'unit' type), but a plain string.
Represents a channel containing a data stream recorded from some source,
such as a neuron, or a stimulus variable.
Unique identifier for the channel.
Sequence number for the channel, out of its group.
Short, precise, descriptive name for the channel.
Units for the channel's values. Note this is not controlled in the way
units generally are in BrainML (based on the BrainMetaL 'unit' type),
but a plain string.