run_tsne: run_tsne

In jrboyd/seqtsne: Application of t-sne to NGS Genomic Signal Profiles

run_tsne

Description

run_tsne

Usage

run_tsne(
  profile_dt,
  perplexity = 100,
  n_cores = getOption("mc.cores", 1),
  high_topright = TRUE,
  norm1 = TRUE,
  Y_init = NULL,
  x_var = "x",
  y_var = "y",
  id_var = "id",
  wide_var = "name",
  tall_var = "tall_none",
  fun.aggregate = mean
)

Arguments

profile_dt	Tidy data.table of profile information. As returned by seqsetvis::ssvFetchBam.
perplexity	perplexity value for t-SNE. Passed to Rtsne::Rtsne.
n_cores	Number of threads for running t-SNE.
high_topright	If TRUE, flip tx/ty as needed such that most points are in the top-right quadrant.
norm1	If TRUE, rescale tx and ty to be centered at 0,0 with total size of 1.
Y_init	Optional initial coordinates for tx and ty.
wide_var	Wide variable. Spreads matrix into columns. Each regions tSNE position will be based on one profile for every value of wide_var.
tall_var	Tall variable. Repeats matrix entries down rows. Each region will appear once per value of tall_var in final tSNE.

Value

Return data.table mapping each id_var entry to tSNE space: defined by tx and ty

Examples

data("profile_dt")
setalloccol(profile_dt)
chiptsne:::run_tsne(profile_dt, wide_var = "wide_var", tall_var = "tall_var")

jrboyd/seqtsne documentation built on Nov. 5, 2022, 6:37 a.m.