show.velocity.on.embedding.eu.Rd

% Generated by roxygen2: do not edit by hand
% Please edit documentation in R/momentum_routines.R
\name{show.velocity.on.embedding.eu}
\alias{show.velocity.on.embedding.eu}
\title{Visualize RNA velocities on an existing embedding using Euclidean-based transition probability matrix within the kNN graph.}
\usage{
show.velocity.on.embedding.eu(emb, vel, n = 30, embedding.knn = TRUE,
  cell.colors = NULL, sigma = NA, beta = 1, arrow.scale = 1,
  scale = "log", nPcs = NA, arrow.lwd = 1, xlab = "", ylab = "",
  control.for.neighborhood.density = TRUE, ntop.trajectories = 1,
  do.par = T, show.cell.arrows = NULL, show.cell.trajectories = NULL,
  show.trajectories = FALSE, show.all.trajectories = FALSE,
  show.cell.diffusion.posterior = NULL, show.grid.flow = FALSE,
  diffusion.steps = 10, cell.dist = NULL,
  trajectory.spline.shape = 1, cell.color.alpha = 0.5,
  n.cores = defaultNCores(), n.trajectory.clusters = 10, ...)
}
\arguments{
\item{emb}{embedding to be used for projection}

\item{vel}{velocity result}

\item{n}{neighborhood size (default=30)}

\item{embedding.knn}{pre-calculated kNN}

\item{cell.colors}{named color vector for cell plotting}

\item{sigma}{sigma to use in calculating transition probability from the eucledian distance (estimated automatically by default)}

\item{beta}{beta parameter used in calculation of transition probability (by default=1)}

\item{arrow.scale}{additional scaling factor for the arrows (default=1)}

\item{scale}{scale to use in calculating distances (default: 'log', also supported 'sqrt'}

\item{nPcs}{number of PCs to project the cells onto (to perform distance calculations in lower dimensions), default=NA which turns off PCA dimensional reduction}

\item{arrow.lwd}{arrow line width}

\item{xlab}{x axis label}

\item{ylab}{y axis label}

\item{control.for.neighborhood.density}{compensate for cell density variations in the embedding (default: TRUE)}

\item{ntop.trajectories}{number of top trajectories to trace back for a given cell (when show.trajectories=TRUE)}

\item{do.par}{whether to reset plotting parameters (default=TRUE)}

\item{show.cell.arrows}{show detailed velocity projection for the specified cell}

\item{show.cell.trajectories}{show trajectories for a specified cell}

\item{show.trajectories}{show top median diffusion trajectories}

\item{show.all.trajectories}{show all diffusion paths (messy)}

\item{show.cell.diffusion.posterior}{show diffusion posterior of a given cell}

\item{show.grid.flow}{show velocity projections on a grid}

\item{diffusion.steps}{number of diffusion steps to take forward (default=10)}

\item{cell.dist}{- optional custom distance (must include all of the cells that are intersecting between emb and vel)}

\item{trajectory.spline.shape}{shape parameter for smoothing median cell trajectories (default=1)}

\item{cell.color.alpha}{trasparency parameter to apply when showing cell colors}

\item{n.cores}{number of cores to use in calculations}

\item{n.trajectory.clusters}{number of trajectory clusters to show median paths for (when show.trajectories=TRUE)}

\item{...}{extra parameters are passed to the plot() function}
}
\value{
transition probability matrix
}
\description{
based on Euclidean distance of the extrapolated cell to others
The direction of the arrow is towards n closest neighbors. The magnitude of the arrow is determined by the cosine projection of the velocity on to the chosen direction
n=1 will only show arrows for cells that end up being projected closer to some other cell than to the original position
n=k (k>1) will show an average direction
Given an expression distance between cells d, and ratio of extrapolated to current expression distances between cells f, the transition probability is calculated as exp(- (d*(f^beta))^2/(2*sigma^2) )
}