swh:1:snp:0da231f3ffdb3226650880f1b61d5d5cdcbd749b
Tip revision: 656fc8b562d53e5d0cedda9e09d9dda81e8c00e9 authored by David Collins on 29 February 2024, 13:44:33 UTC
Merge pull request #8550 from satijalab/develop
Merge pull request #8550 from satijalab/develop
Tip revision: 656fc8b
ProjectIntegration.Rd
% Generated by roxygen2: do not edit by hand
% Please edit documentation in R/integration.R
\name{ProjectIntegration}
\alias{ProjectIntegration}
\title{Integrate embeddings from the integrated sketched.assay}
\usage{
ProjectIntegration(
object,
sketched.assay = "sketch",
assay = "RNA",
reduction = "integrated_dr",
features = NULL,
layers = "data",
reduction.name = NULL,
reduction.key = NULL,
method = c("sketch", "data"),
ratio = 0.8,
sketched.layers = NULL,
seed = 123,
verbose = TRUE
)
}
\arguments{
\item{object}{A Seurat object with all cells for one dataset}
\item{sketched.assay}{Assay name for sketched-cell expression (default is 'sketch')}
\item{assay}{Assay name for original expression (default is 'RNA')}
\item{reduction}{Dimensional reduction name for batch-corrected embeddings
in the sketched object (default is 'integrated_dr')}
\item{features}{Features used for atomic sketch integration}
\item{layers}{Names of layers for correction.}
\item{reduction.name}{Name to save new reduction as; defaults to
\code{paste0(reduction, '.orig')}}
\item{reduction.key}{Key for new dimensional reduction; defaults to creating
one from \code{reduction.name}}
\item{method}{Methods to construct sketch-cell representation
for all cells (default is 'sketch'). Can be one of:
\itemize{
\item \dQuote{\code{sketch}}: Use random sketched data slot
\item \dQuote{\code{data}}: Use data slot
}}
\item{ratio}{Sketch ratio of data slot when \code{dictionary.method} is set
to \dQuote{\code{sketch}}; defaults to 0.8}
\item{sketched.layers}{Names of sketched layers, defaults to all
layers of \dQuote{\code{object[[assay]]}}}
\item{seed}{A positive integer. The seed for the random number generator, defaults to 123.}
\item{verbose}{Print progress and message}
}
\value{
Returns a Seurat object with an integrated dimensional reduction
}
\description{
The main steps of this procedure are outlined below. For a more detailed
description of the methodology, please see Hao, et al Biorxiv 2022:
\doi{10.1101/2022.02.24.481684}
}
\details{
First learn a atom dictionary representation to reconstruct each cell.
Then, using this dictionary representation,
reconstruct the embeddings of each cell from the integrated atoms.
}