lisa_test

Computes LISA statistics for identifying local clusters and outliers.
Used to filter genes driven by only a few cells.

internal

A unified framework for detecting spatially variable genes (SVGs)
in spatial transcriptomics data. This package integrates multiple
state-of-the-art SVG detection methods including 'MERINGUE' (Moran's I based
spatial autocorrelation), 'Giotto' binSpect (binary spatial enrichment test),
'SPARK-X' (non-parametric kernel-based test), and 'nnSVG' (nearest-neighbor
Gaussian processes). Each method is implemented with optimized performance
through vectorization, parallelization, and 'C++' acceleration where applicable.
Methods are described in Miller et al. (2021) <doi:10.1101/gr.271288.120>,
Dries et al. (2021) <doi:10.1186/s13059-021-02286-2>,
Zhu et al. (2021) <doi:10.1186/s13059-021-02404-0>, and
Weber et al. (2023) <doi:10.1038/s41467-023-39748-z>.

Zaoqu Liu

Spatially Variable Genes Detection Methods for Spatial
Transcriptomics

lisa_test function

<dl><dt>x</dt>
<dd>Numeric vector of values.</dd>
<dt>W</dt>
<dd>Spatial weights matrix.</dd>
<dt>alternative</dt>
<dd>Alternative hypothesis ("greater", "less", "two.sided").</dd></dl>

Arguments

Local Indicators of Spatial Association (LISA) — lisa_test

<dl>

<dt>x</dt>
<dd>Numeric vector of values.</dd>


<dt>W</dt>
<dd>Spatial weights matrix.</dd>


<dt>alternative</dt>
<dd>Alternative hypothesis ("greater", "less", "two.sided").</dd>

</dl>

lisa_test: Local Indicators of Spatial Association (LISA)

Description

Usage

Value

Arguments