Skip to contents

Fetch neuprint metadata for MANC neurons

Source: R/neuprint.R
manc_neuprint_meta.Rd

Fetch neuprint metadata for MANC neurons

Usage

manc_neuprint_meta(
  ids = NULL,
  conn = manc_neuprint(),
  cache = NA,
  roiInfo = FALSE,
  fields.regex.exclude = NULL,
  fields.regex.include = NULL,
  ...
)

Arguments

ids

A set of body ids in any form understandable to manc_ids

conn

Optional, a neuprint_connection object, which also specifies the neuPrint server. Defaults to manc_neuprint() to ensure that query is against the VNC dataset.

cache

whether to cache the query. When cache=NA (the default) queries are cached for neuprint snapshot versions (but not production datasets). See details.

roiInfo

whether to include the roiInfo field detailing synapse numbers in different locations. This is omitted by default as it is returned as a character vector of unprocessed JSON.

fields.regex.exclude, fields.regex.include

Optional regular expressions to define fields to include or exclude from the returned metadata.

...

Additional arguments passed to neuprint_get_meta

Value

A data.frame with one row for each (unique) id and NAs for all columns except bodyid when neuprint holds no metadata.

Details

When ids = NULL then a default set of bodies is selected using the manc_dvid_annotations function. Since April 2025 this uses the node='neuprint'. This should correspond to all neurons with an annotation. You can also use other searches e.g. to fetch all neurons, see examples.

This function will now cache neuprint queries when using a snapshot release (which is assumed not to change). Snapshot releases are identified by containing the string ":v" as in manc:v1.2.3. The cache currently lasts for 24h.

See also

manc_ids

Examples

# \donttest{
manc_neuprint_meta("DNp01")
#> Error in neuprint_login(server = server, dataset = dataset, token = token,     ...): Sorry you must specify a neuprint server! See ?neuprint_login for details!

# use of a full CYPHER query. NB Each field relating to the neuron must be
# be preceded by "n."
bignogroup <-
  manc_neuprint_meta("where:NOT exists(n.group) AND n.synweight>5000 AND n.class CONTAINS 'neuron'")
#> Error in neuprint_login(server = server, dataset = dataset, token = token,     ...): Sorry you must specify a neuprint server! See ?neuprint_login for details!
head(bignogroup)
#> Error: object 'bignogroup' not found
# }
if (FALSE) { # \dontrun{
# fetch all neurons
allneurons <- manc_neuprint_meta('where:exists(n.bodyId)')
# in theory you could do this, but it often seems to time out:
allsegs=neuprintr::neuprint_ids('where:exists(n.bodyId)', all_segments=TRUE)
} # }