calculateEvalMetrics

Calculate evaluation metrics on test mixed transcriptional profiles. By
default, absolute error (<code>AbsErr</code>), proportional absolute error
(<code>ppAbsErr</code>), squared error (<code>SqrErr</code>), and proportional squared
error (<code>ppSqrErr</code>) are calculated for each test mixed profile. In
addition, each of these metrics is aggregated according to three criteria:
cell type (<code>CellType</code>), probability bins in ranges of 0.1 (<code>pBin</code>),
and number of different cell types present in the spot (<code>nCellTypes</code>).

Deconvolution of spatial transcriptomics data based on neural networks and single-cell RNA-seq data. SpatialDDLS implements a workflow to create neural network models able to make accurate estimates of cell composition of spots from spatial transcriptomics data using deep learning and the meaningful information provided by single-cell RNA-seq data. See Torroja and Sanchez-Cabo (2019) <doi:10.3389/fgene.2019.00978> and Mañanes et al. (2024) <doi:10.1093/bioinformatics/btae072> to get an overview of the method and see some examples of its performance.

Diego Ma<c3><b1>anes

SpatialDDLS

Deconvolution of Spatial Transcriptomics Data Based on Neural
Networks

Diego Mañanes

Carlos Torroja

Fatima Sanchez-Cabo

calculateEvalMetrics function

<dl><dt>object</dt>
<dd><code>SpatialDDLS</code> object with a trained model in
<code>trained.model</code> slot and the actual cell proportions of test mixed
profiles in <code>prob.cell.types</code> slot.</dd></dl>

Arguments

Calculate evaluation metrics on test mixed transcriptional profiles — calculateEvalMetrics

<dl>

<dt>object</dt>
<dd><code>SpatialDDLS</code> object with a trained model in
<code>trained.model</code> slot and the actual cell proportions of test mixed
profiles in <code>prob.cell.types</code> slot.</dd>

</dl>

calculateEvalMetrics: Calculate evaluation metrics on test mixed transcriptional profiles

Description

Usage

Value

Arguments

See Also