Cosine plot

Usage

mcns_cosine_plot(
  ids,
  partners = c("output", "input"),
  group = FALSE,
  groupfun = NULL,
  labRow = "{name}_{group}",
  predict.manc = FALSE,
  heatmap = TRUE,
  metadata.source = c("neuprint", "clio"),
  interactive = FALSE,
  action = NULL,
  threshold = 10,
  ...
)

Arguments

ids: Bodyids in any form understood by mcns_ids
partners: Whether to cluster based on connections to input or output partner neurons (default both).
group: Optional character vector specifying the grouping column for the partner neurons when constructing the cosine similarity matrix. TRUE implies to 'group'.
groupfun: A function which receives the metadata for all partner neurons and returns a single grouping vector (see the details section).
labRow: A string interpolated by glue using the dataframe of metadata fetched by mcns_neuprint_meta. Alternatively a character vector with as many elements as ids directly specifying the rows.
predict.manc: whether to use manc_bodyid predicted matches to define grouping information in addition to curated manc_group matches. See mcns_predict_group for details.
heatmap: A logical indicating whether or not to plot the heatmap OR a function to plot the heatmap whose argument names are compatible with stats::heatmap. gplots::heatmap.2 is a good example. Defaults to TRUE therefore plotting the full heatmap with stats::heatmap.
metadata.source: Whether to use neuprint (mcns_neuprint_meta) and clio ()
interactive: Whether to plot an interactive heatmap (allowing zooming and id selection). See details.
action: Whether to zero out or drop any NA values in the cosine matrix (these may be present when some columns have no entries)
threshold: An integer threshold (connections >= this will be returned)
...: additional arguments passed to neuprint_cosine_plot and eventually to heatmap.

Value

The result of heatmap invisibly including the row and column dendrograms.

Examples

# \donttest{
# cosine clustering based on grouped output partners (mainly DNs right now)
r=mcns_cosine_plot("/name:LAL.+", partners='out', group=TRUE)
#> Warning: Dropping: 93/24092 neurons representing 1646/862740 synapses due to missing ids!
#> Warning: diag(V) has non-positive or non-finite entries; finite result is doubtful

# }
if (FALSE) { # \dontrun{
# fancier labelling of rows including soma side of neurons
r=mcns_cosine_plot("/name:LAL.+", partners='out', group=TRUE,
  labRow = '{name}_{group}_{mcns_soma_side(data.frame(bodyid, name, somaLocation))}')

r2=mcns_cosine_plot("/name:LAL.+", partners='out', group=TRUE,
  labRow = '{instance}_{group}_{soma_side}', metadata.source='clio')

# interactive version (open in browser)
mcns_cosine_plot('/name:Pm2.*R', group=T, interactive = T)

# just return an hclust (dendrogram) object without plotting anything
pm2.hc=mcns_cosine_plot('/name:Pm2.*R', group=T, heatmap=FALSE)
plot(pm2.hc)
} # }