02_child_collapse_replicates.Rmd

<!--
Child document for 01_normDiffana.Rmd.
The environment is inherited from the parent file.
-->

# Collapsing technical replicates

The aim of this section is to pool the technical replicates. The raw counts for samples bearing the same value in the `RepTechGroup` column are summed.

```{r dl_02, echo = FALSE, results = "asis"}
dl_file = "./02_child_collapse_replicates.Rmd"

cat(sprintf('The code behind this section can be downloaded at:
    <a download="%s" href="data:%s;base64,%s">
    <button class="btn-custom">Collapse Replicates file (Rmd)</button>
    </a>',
    basename(dl_file),
    "text/csv",
    base64enc::base64encode(dl_file)))
```

## Data handling

We collapse the technical replicates by summing their raw count for each gene. We display the top left corner of this matrix, meaning 5 genes and 5 samples. As for the raw counts, collapsed count data are integer values.

```{r collapse_replicates_deseq2}
dds = DESeq2::collapseReplicates(dds,
                                 groupby = dds$RepTechGroup)

DESeq2::counts(dds)[c(1:5), c(1:5)] %>%
  knitr::kable("html") %>%
  kableExtra::kable_styling(full_width = FALSE)
```

We update the design matrix accordingly.

```{r collapse_design}
design = design %>%
  dplyr::filter(!duplicated(RepTechGroup)) %>%
  base::droplevels()
```

We save the pooled count matrix as a TSV file.

```{r save_pooled_count_matrix}
filename = paste0(prefix, projectName,
                  "-normalisation_rawPooledCountMatrix.tsv")

saveTable(DESeq2::counts(dds),
          fileName = filename)
```

Output file name:<br>*`r filename`*

## Visualisation

### Pooled counts barplot

This figure displays the total number of reads per sample, where technical replicates have been pooled.

```{r fig-normalisation_barplot_counts_pooled, fig.width = 10, fig.height = 6}
colSums(DESeq2::counts(dds)) %>%
  data.frame(sample = factor(names(.), levels = levels(design$RepTechGroup)),
             read_counts = .) %>%
  dplyr::left_join(x = .,
                   y = design[, c("RepTechGroup", "Condition")],
                   by = c("sample" = "RepTechGroup")) %>%
  # Plot
  plotBarplot(.,
              y_column = "read_counts",
              y_title = "Total read counts",
              fig_title = paste0("Pooled read counts - ", projectName))
```

### Pooled counts boxplot

This figure displays, as a boxplot, the number of counts for each gene, per sample, where technical replicates have been pooled.

```{r fig-normalisation_boxplot_count_pooled, fig.width = 10, fig.height = 6}
log2(DESeq2::counts(dds)+1) %>%
  reshape2::melt() %>%
  `colnames<-`(c("gene_ID", "sample", "log2_counts")) %>%
  dplyr::mutate(sample = factor(sample, levels = levels(design$RepTechGroup))) %>%
  dplyr::left_join(x = .,
                   y = unique(design[, c("RepTechGroup", "Condition")]),
                   by = c("sample" = "RepTechGroup")) %>%
  # Plot
  plotBoxplot(.,
              y_column = "log2_counts",
              y_title = "log_2 (counts+1)",
              fig_title = paste0("Pooled count distribution - ", projectName))
```

------------------------------------------------------------------------

<!--
End of the child file.
Coming back to the parent file.
-->