Register Aedes dataset for coconatfly — register_aedes

Usage

register_aedes_coconat(showerror = TRUE)

Arguments

showerror: Logical; when FALSE, return invisibly if dependencies are missing.

Value

Invisible NULL.

Details

The aedes dataset is continually evolving. You have three main choices for how to handle this.

use a specific numeric version (aka materialisation) of the segmentation.
use the latest materialisation version (version='latest')
map ids to the current time (version='now')

Option 3 is the default (see aedes_set_version()). Option 2 can make queries more stable, but note that 'latest' can be several days old.

Metadata returned for aedes neurons includes a pgroup column from aedes_predict_group(), which can be used to group partner neurons when clustering by connectivity, e.g. cf_cosine_plot(ids, group = "pgroup"). See vignette("connectivity-clustering") for a worked example.

Examples

if (FALSE) { # \dontrun{
register_aedes_coconat()
cf_meta(cf_ids(aedes="/class:MBON.*"))
aedes_set_version('now')
cf_meta(cf_ids(aedes="/class:MBON.*"))

# cluster a group of neurons by connectivity, grouping partner neurons by
# their curated group (partners without one are dropped) ...
cf_cosine_plot(cf_ids(aedes = "group:36155"), group = "group",
  labRow = "{side}_{serial_id}")
# ... or by predicted group (type, then group, then nblast cluster), so that
# far fewer partners are lost
cf_cosine_plot(cf_ids(aedes = "group:36155"), group = "pgroup",
  labRow = "{side}_{serial_id}")
} # }