> Status: `draft`
>
> Template class: SOURCE-BACKED WORKFLOW

## Purpose

Normalize a filtered RNA Seurat object with Seurat log normalization, identify
variable features, scale the selected assay, and run PCA for downstream use.

## Inputs and parameters

```{r}
library(Seurat)
library(qs2)
library(ggplot2)

input_object <- "input/seurat_object_qc.qs2"
output_object <- "output/seurat_object_log_normalized.qs2"
output_plot <- "output/variable_features.png"

assay_name <- "RNA"
scale_factor <- 10000
variable_feature_method <- "vst"
n_variable_features <- 3000
vars_to_regress <- NULL
n_pcs <- 30
pca_reduction <- "pca"
```

## Analysis

```{r}
object <- qs2::qs_read(input_object)
DefaultAssay(object) <- assay_name

object <- NormalizeData(
  object,
  assay = assay_name,
  normalization.method = "LogNormalize",
  scale.factor = scale_factor,
  verbose = FALSE
)
object <- FindVariableFeatures(
  object,
  assay = assay_name,
  selection.method = variable_feature_method,
  nfeatures = n_variable_features,
  verbose = FALSE
)

object <- ScaleData(
  object,
  assay = assay_name,
  features = VariableFeatures(object),
  vars.to.regress = vars_to_regress,
  verbose = FALSE
)
object <- RunPCA(
  object,
  assay = assay_name,
  features = VariableFeatures(object),
  npcs = n_pcs,
  reduction.name = pca_reduction,
  verbose = FALSE
)
```

## Diagnostics

```{r}
dir.create(dirname(output_plot), recursive = TRUE, showWarnings = FALSE)
print(head(VariableFeatures(object), 20))
print(length(VariableFeatures(object)))
p <- VariableFeaturePlot(object, assay = assay_name)
ggsave(output_plot, p, width = 7, height = 5, dpi = 150)
```

## Outputs

```{r}
dir.create(dirname(output_object), recursive = TRUE, showWarnings = FALSE)
qs2::qs_save(object, output_object)
```

## Method notes

This is the baseline log-normalized route. The scale factor, feature
selection, regression variables, and number of PCs affect downstream analysis.
The output is PCA-ready: it contains normalized data, selected variable
features, scaled data, and the named PCA reduction.
SCTransform is a separate method, not a hidden branch here.