In bakerwm/goldclipReport: Reporter for GoldCLIP (Title Case)

suppressPackageStartupMessages(library(goldclipReport))
suppressPackageStartupMessages(library(dplyr))
suppressPackageStartupMessages(library(knitr))
suppressPackageStartupMessages(library(kableExtra))

knitr::opts_chunk$set(fig.width = 12, 
                      fig.height = 8, 
                      fig.path = 'Figs/',
                      echo = FALSE,
                      eval = TRUE,
                      cache = FALSE,
                      prompt = FALSE,
                      tidy = FALSE,
                      comment = NA,
                      message = FALSE,
                      warning = FALSE,
                      rownames.print = FALSE)
options(width=150)

print_df <- function(x){
  x %>%
  kableExtra::kable() %>%
  kableExtra::kable_styling() %>%
  kableExtra::scroll_box(width = "100%", height = "400px")
}

demx_path <- params$demx_path

1. Summary

# demx_path  <- "/nas/yulab/seq_data/Yu_2019/YY35_20190325/results/"
# demx_path  <- "/nas/yulab/seq_data/Yu_2019/YY39_20190528/results"
df         <- demxParser(demx_path)
total      <- sum(df$count)
undemx     <- df %>% 
  dplyr::filter(id == "undemx") %>% 
  pull(count)
undemx_pct <- round((undemx / total) * 100, 2)
demx       <- total - undemx
demx_pct   <- 100 - undemx_pct
n_samples  <- nrow(df) - 1
total      <- scales::number(total, big.mark = ",")
demx       <- scales::number(demx, big.mark = ",")
undemx     <- scales::number(undemx, big.mark = ",")

summary_text <- glue::glue("{total} reads were generated from this Lane; \n{demx} ({demx_pct}%)  of the reads were demultiplexed for {n_samples} samples; \n{undemx} ({undemx_pct}%) of the reads were not assigned for any of the samples. ")

print(summary_text)

2. Number of reads

df_print <- df %>%
  tibble::rowid_to_column("num") %>%
  dplyr::mutate(count = scales::number(count, big.mark = ","))

# print_df(df_print)
kable(df_print) %>%
  kable_styling(bootstrap_options = c("striped", "hover", "condensed"))

3. Plot

p <- numberBarplot(df)
print(p)

4. Bacteria content

In order to check the bacteria content in each sample, Kraken2 was used to map sequencing reads to NCBI database: (bacteria, Archaea, Virus and Homo sapiens).

The number of reads for each category were normalized by RPM, log10(RPM) were shown in the following figure.

bacteria_csv <- file.path(demx_path, "bacteria", "bacteria_content.csv")

df <- read.csv(bacteria_csv) %>%
  tibble::column_to_rownames("X")

ma <- log10(df + 1)

## make plot
cc = colorRampPalette(rev(RColorBrewer::brewer.pal(n = 7,
                                     name = "RdYlBu")))
breaks = seq(0, 6, length.out = 100)

# plot
p <- pheatmap::pheatmap(ma,
              silent = TRUE,
              cluster_cols = FALSE,
              color = cc(100),
              breaks = breaks,
              border_color = "grey40")