> ## Documentation Index
> Fetch the complete documentation index at: https://mintlify.com/satijalab/seurat-wrappers/llms.txt
> Use this file to discover all available pages before exploring further.

# RunALRA()

> Zero-preserving imputation of scRNA-seq data using adaptively-thresholded low-rank approximation

`RunALRA()` implements the ALRA imputation algorithm. It computes a rank-k approximation of the normalized expression matrix, then adaptively thresholds values based on the distribution of negative values in the approximation. Zero values are preserved only when they are truly absent.

## Syntax

```r theme={null}
RunALRA(
  object,
  k = NULL,
  q = 10,
  quantile.prob = 0.001,
  use.mkl = FALSE,
  mkl.seed = -1,
  assay = NULL,
  slot = "data",
  setDefaultAssay = TRUE,
  genes.use = NULL,
  K = NULL,
  p.val.th = 1e-10,
  noise.start = NULL,
  q.k = 2,
  k.only = FALSE,
  ...
)
```

## Parameters

<ParamField path="object" type="Seurat" required>
  A Seurat object. Also accepts a matrix-like object when called as a default method.
</ParamField>

<ParamField path="k" type="integer" default="NULL">
  Rank of the low-rank approximation. Set to `NULL` for automated selection of `k`.
</ParamField>

<ParamField path="q" type="integer" default="10">
  Number of additional power iterations in randomized SVD when computing the rank-k approximation.
</ParamField>

<ParamField path="quantile.prob" type="numeric" default="0.001">
  Quantile probability used to calculate the per-gene threshold.
</ParamField>

<ParamField path="use.mkl" type="logical" default="FALSE">
  Use the Intel MKL-based SVD implementation (requires rpca-mkl package).
</ParamField>

<ParamField path="mkl.seed" type="integer" default="-1">
  Seed for the MKL SVD random generator. Any negative value uses the current timestamp. Only relevant when `use.mkl = TRUE`.
</ParamField>

<ParamField path="assay" type="character" default="NULL">
  Name of the assay to use. Defaults to the active default assay.
</ParamField>

<ParamField path="slot" type="character" default="data">
  Slot within the assay to use as input.
</ParamField>

<ParamField path="setDefaultAssay" type="logical" default="TRUE">
  If `TRUE`, sets the resulting `alra` assay as the default assay after imputation.
</ParamField>

<ParamField path="genes.use" type="character vector" default="NULL">
  Subset of genes to impute. Defaults to all genes in the assay.
</ParamField>

<ParamField path="K" type="integer" default="NULL">
  Number of singular values to compute when choosing `k` automatically. Must be less than the smallest matrix dimension. Defaults to 100 or the smallest dimension, whichever is less.
</ParamField>

<ParamField path="p.val.th" type="numeric" default="1e-10">
  P-value threshold for significance when choosing `k` automatically.
</ParamField>

<ParamField path="noise.start" type="integer" default="NULL">
  Index from which all smaller singular values are considered noise during automatic `k` selection. Defaults to `K - 20`.
</ParamField>

<ParamField path="q.k" type="integer" default="2">
  Number of additional power iterations when choosing `k` automatically.
</ParamField>

<ParamField path="k.only" type="logical" default="FALSE">
  If `TRUE`, computes the optimal `k` and stores it in the object without performing imputation. Use with `ALRAChooseKPlot()` to visualize the choice.
</ParamField>

## Returns

A Seurat object with a new assay named `alra` containing the imputed expression matrix. When `setDefaultAssay = TRUE`, this becomes the default assay. When `k.only = TRUE`, returns the object with the computed `k` stored in the tool slot.

## Examples

```r theme={null}
library(SeuratWrappers)

# Simple one-step imputation
pbmc_small <- RunALRA(object = pbmc_small)

# Two-step: choose k first, then impute
pbmc_alra <- RunALRA(pbmc_small, k.only = TRUE)
ggouts <- ALRAChooseKPlot(pbmc_alra)
do.call(gridExtra::grid.arrange, c(ggouts, nrow = 1))
pbmc_alra <- RunALRA(pbmc_alra)
```

## See Also

* [`ALRAChooseKPlot()`](/api/alra-choose-k-plot) — Visualize singular value spectrum for `k` selection
* [ALRA method guide](/methods/alra)