2023-08-29 qPCR Analysis

Author

Dan Rice

Published

August 29, 2023

Objectives

Compare several wastewater filtering options by measuring the nucleic acid content by qPCR. Experimental design here.
Get Dan up to speed with working with this data.
Explore options for data analysis workflows.

Preliminary work

Ari put the .eds files into Google Drive from the lab computer. (He also exported some Excel files but we’re not using those.)
Dan installed the Google Drive desktop app and Design and Analysis on his Mac, opened the .eds files, and fixed some missing data in the plate layout
Dan used the “Analyze” in Design and Analysis to automatically calculate thresholds and compute c_q values.
Dan exported the data to .csv.
Symlinked the google drive folder for the airport experiment to ~/airport/ on his computer so I don’t have to refer to the whole filepath here.

Data import

library(readr)
library(dplyr)


Attaching package: 'dplyr'

The following objects are masked from 'package:stats':

    filter, lag

The following objects are masked from 'package:base':

    intersect, setdiff, setequal, union

library(purrr)

data_dir <- "~/airport/[2023-08-29] CP Prefilters vs Vacuum Filters/Test 1 qPCR Results/csv/"
filename_pattern <- "Results"

raw_data <- list.files(
  data_dir,
  pattern = filename_pattern,
  full.names = TRUE
) |>
  map(function(f) read_csv(f, skip = 23)) |>
  list_rbind()

Rows: 36 Columns: 21
── Column specification ────────────────────────────────────────────────────────
Delimiter: ","
chr (8): Well Position, Sample, Target, Task, Reporter, Quencher, Amp Status...
dbl (8): Well, Amp Score, Cq Confidence, Cq Mean, Cq SD, Threshold, Baseline...
lgl (5): Omit, Curve Quality, Result Quality Issues, Auto Threshold, Auto Ba...

ℹ Use `spec()` to retrieve the full column specification for this data.
ℹ Specify the column types or set `show_col_types = FALSE` to quiet this message.
Rows: 29 Columns: 21
── Column specification ────────────────────────────────────────────────────────
Delimiter: ","
chr (8): Well Position, Sample, Target, Task, Reporter, Quencher, Amp Status...
dbl (8): Well, Amp Score, Cq Confidence, Cq Mean, Cq SD, Threshold, Baseline...
lgl (5): Omit, Curve Quality, Result Quality Issues, Auto Threshold, Auto Ba...

ℹ Use `spec()` to retrieve the full column specification for this data.
ℹ Specify the column types or set `show_col_types = FALSE` to quiet this message.

print(raw_data)

# A tibble: 65 × 21
    Well `Well Position` Omit  Sample Target Task    Reporter Quencher
   <dbl> <chr>           <lgl> <chr>  <chr>  <chr>   <chr>    <chr>   
 1     1 A1              FALSE 1      CrA    UNKNOWN FAM      NFQ-MGB 
 2     2 A2              FALSE 1      CrA    UNKNOWN FAM      NFQ-MGB 
 3     3 A3              FALSE 1      CrA    UNKNOWN FAM      NFQ-MGB 
 4     5 A5              FALSE 1      16S    UNKNOWN FAM      NFQ-MGB 
 5     6 A6              FALSE 1      16S    UNKNOWN FAM      NFQ-MGB 
 6     7 A7              FALSE 1      16S    UNKNOWN FAM      NFQ-MGB 
 7    13 B1              FALSE 2      CrA    UNKNOWN FAM      NFQ-MGB 
 8    14 B2              FALSE 2      CrA    UNKNOWN FAM      NFQ-MGB 
 9    15 B3              FALSE 2      CrA    UNKNOWN FAM      NFQ-MGB 
10    17 B5              FALSE 2      16S    UNKNOWN FAM      NFQ-MGB 
# ℹ 55 more rows
# ℹ 13 more variables: `Amp Status` <chr>, `Amp Score` <dbl>,
#   `Curve Quality` <lgl>, `Result Quality Issues` <lgl>, Cq <chr>,
#   `Cq Confidence` <dbl>, `Cq Mean` <dbl>, `Cq SD` <dbl>,
#   `Auto Threshold` <lgl>, Threshold <dbl>, `Auto Baseline` <lgl>,
#   `Baseline Start` <dbl>, `Baseline End` <dbl>

coding <- list(
  "1" = "Normal centrifugation/filtration, regular CP",
  "2" = "Normal centrifugation/filtration, regular CP",
  "3" = "Normal centrifugation, no filtration, prefilter CP",
  "4" = "Normal centrifugation, no filtration, prefilter CP",
  "5" = "No centrifugation/filtration, prefilter CP",
  "6" = "No centrifugation/filtration, prefilter CP"
)
tidy_data <- raw_data |>
  mutate(
    Cq = as.double(Cq),
    Treatment = recode(Sample, !!!coding)
  )

Warning: There was 1 warning in `mutate()`.
ℹ In argument: `Cq = as.double(Cq)`.
Caused by warning:
! NAs introduced by coercion

library(ggplot2)
tidy_data |>
  ggplot(mapping = aes(x = Cq, y = Treatment, color = Sample)) +
  geom_point(alpha = 0.5) +
  facet_wrap(facets = ~Target)

Warning: Removed 11 rows containing missing values (`geom_point()`).

TODO

~~Figure out what the Sample numbers mean with respect to the different treatments~~
There are a few NaNs that show up as missing points and aren’t evident in the plot. Will investigate later.