Skip to contents

Get all the modifications associated with one or more DVID nodes

Usage

manc_mutations(nodes = "neutu", include_first = NA, bigcols = FALSE, ...)

Arguments

nodes

One or more DVID nodes. Ranges can be specified as "first:last". The special value "all" implies the full sequence from the root node.

include_first

Whether to include the first node in a sequence. The default behaviour (when =NA) will ignore the first node in a range specified first:last but keep the root nodes for the special range of "all".

bigcols

Whether to include big columns in the results. The CleavedSupervoxels column in particular is very large and probably is not that useful for many. Default FALSE.

...

Additional arguments passed to pbapply

Value

A tibble with columns including:

  • Action: merge, cleave, supervoxel-split

  • App: Typically NeuTu/Neu3

  • Target: For NeuTu the bodyid of the (larger) target object

  • Labels: For NeuTu the bodyids of the (smaller) objects being merged into the (larger) target object.

  • CleavedLabel: For Neu3 the label of the new smaller cleaved body

  • OrigLabel: For Neu3 the label of the larger cleaved body (which should retain its id)

  • Timestamp: Absolute time in UTC at which change was committed in POSIXct format.

  • Reltimestamp: Relative timestamp (in seconds) referenced to the time at which new DVID node was opened.

Examples

if (FALSE) { # \dontrun{
library(ggplot2)
library(dplyr)
# find all mutations in neutu but not yet in clio
notin_neuprint=manc_mutations('neuprint:neutu')
subset(notin_neuprint, User!='bergs') %>%
  qplot(Timestamp, data=., bins=100, fill=Action)

library(ggplot2)

allmuts=manc_mutations('all')
qplot(Timestamp, fill=App, data=allmuts, bins=100)

allmuts %>%
  filter(User!='bergs') %>%
  qplot(Timestamp, data=., bins=100, fill=App)

subset(allmuts, User!='bergs') %>%
  qplot(Timestamp, data=., bins=100, fill=Action)
} # }