R Notebook

library(tidyverse)

x2016_04_01_a549_48hr_batch1_cmap_style_SQ00015233_normalized <- 
  read_csv("~/work/projects/2015_10_05_DrugRepurposing_AravindSubramanian_GolubLab_Broad/workspace/backend/2016_04_01_a549_48hr_batch1_cmap_style/SQ00015233/SQ00015233_normalized.csv")

Parsed with column specification:
cols(
  .default = col_double(),
  Metadata_Plate = col_character(),
  Metadata_Well = col_character(),
  Metadata_Assay_Plate_Barcode = col_character(),
  Metadata_Plate_Map_Name = col_character(),
  Metadata_Batch_Date = col_date(format = ""),
  Metadata_well_position = col_character(),
  Metadata_broad_sample = col_character(),
  Metadata_solvent = col_character(),
  Metadata_pert_id = col_character(),
  Metadata_pert_mfc_id = col_character(),
  Metadata_pert_well = col_character(),
  Metadata_pert_id_vendor = col_logical(),
  Metadata_cell_id = col_character(),
  Metadata_broad_sample_type = col_character(),
  Metadata_pert_vehicle = col_character(),
  Metadata_pert_type = col_character(),
  Metadata_pert_iname = col_character(),
  Metadata_moa = col_character(),
  Metadata_pert_mfc_desc = col_character(),
  Metadata_pert_name = col_character()
)
See spec(...) for full column specifications.

x2016_04_01_a549_48hr_batch1_SQ00015233_normalized <- 
  read_csv("~/work/projects/2015_10_05_DrugRepurposing_AravindSubramanian_GolubLab_Broad/workspace/backend/2016_04_01_a549_48hr_batch1/SQ00015233/SQ00015233_normalized.csv")

Parsed with column specification:
cols(
  .default = col_double(),
  Metadata_Plate = col_character(),
  Metadata_Well = col_character(),
  Metadata_Assay_Plate_Barcode = col_character(),
  Metadata_Plate_Map_Name = col_character(),
  Metadata_Batch_Date = col_date(format = ""),
  Metadata_well_position = col_character(),
  Metadata_broad_sample = col_character(),
  Metadata_solvent = col_character(),
  Metadata_pert_id = col_character(),
  Metadata_pert_mfc_id = col_character(),
  Metadata_pert_well = col_character(),
  Metadata_pert_id_vendor = col_logical(),
  Metadata_cell_id = col_character(),
  Metadata_broad_sample_type = col_character(),
  Metadata_pert_vehicle = col_character(),
  Metadata_pert_type = col_character(),
  Metadata_pert_iname = col_character(),
  Metadata_moa = col_character(),
  Metadata_pert_mfc_desc = col_character(),
  Metadata_pert_name = col_character()
)
See spec(...) for full column specifications.

variables <- str_subset(names(x2016_04_01_a549_48hr_batch1_cmap_style_SQ00015233_normalized), "Cells|Nuclei|Cytoplasm")

x2016_04_01_a549_48hr_batch1_cmap_style_SQ00015233_normalized_mean_full_plate <-
  x2016_04_01_a549_48hr_batch1_cmap_style_SQ00015233_normalized %>%
  cytominer::aggregate(variables,
                       c("Metadata_Plate"),
                       operation = "mean",
                       na.rm = TRUE) %>%
  pivot_longer(-Metadata_Plate) %>%
  na.omit()

x2016_04_01_a549_48hr_batch1_cmap_style_SQ00015233_normalized_median_full_plate <-
  x2016_04_01_a549_48hr_batch1_cmap_style_SQ00015233_normalized %>%
  cytominer::aggregate(variables,
                       c("Metadata_Plate"),
                       operation = "median",
                       na.rm = TRUE) %>%
  pivot_longer(-Metadata_Plate) %>%
  na.omit()

x2016_04_01_a549_48hr_batch1_cmap_style_SQ00015233_normalized_mean_dmso <-
  x2016_04_01_a549_48hr_batch1_cmap_style_SQ00015233_normalized %>%
  filter(Metadata_broad_sample_type == "control") %>%
  cytominer::aggregate(variables,
                       c("Metadata_Plate"),
                       operation = "mean",
                       na.rm = TRUE) %>%
  pivot_longer(-Metadata_Plate) %>%
  na.omit()

x2016_04_01_a549_48hr_batch1_cmap_style_SQ00015233_normalized_median_dmso <-
  x2016_04_01_a549_48hr_batch1_cmap_style_SQ00015233_normalized %>%
  filter(Metadata_broad_sample_type == "control") %>%
  cytominer::aggregate(variables,
                       c("Metadata_Plate"),
                       operation = "median",
                       na.rm = TRUE) %>%
  pivot_longer(-Metadata_Plate) %>%
  na.omit()

x2016_04_01_a549_48hr_batch1_cmap_style_SQ00015233_normalized_mean_full_plate %>%
  ggplot(aes(value)) + 
  geom_histogram() +
  ggtitle("x2016_04_01_a549_48hr_batch1_cmap_style_SQ00015233_normalized_mean_full_plate")

x2016_04_01_a549_48hr_batch1_cmap_style_SQ00015233_normalized_median_full_plate %>%
  ggplot(aes(value)) + 
  geom_histogram() +
  ggtitle("x2016_04_01_a549_48hr_batch1_cmap_style_SQ00015233_normalized_median_full_plate")

x2016_04_01_a549_48hr_batch1_cmap_style_SQ00015233_normalized_mean_dmso %>%
  ggplot(aes(value)) + 
  geom_histogram() +
  ggtitle("x2016_04_01_a549_48hr_batch1_cmap_style_SQ00015233_normalized_mean_dmso")

x2016_04_01_a549_48hr_batch1_cmap_style_SQ00015233_normalized_median_dmso %>%
  ggplot(aes(value)) + 
  geom_histogram() +
  ggtitle("x2016_04_01_a549_48hr_batch1_cmap_style_SQ00015233_normalized_median_dmso")

x2016_04_01_a549_48hr_batch1_SQ00015233_normalized_mean_full_plate <-
  x2016_04_01_a549_48hr_batch1_SQ00015233_normalized %>%
  cytominer::aggregate(variables,
                       c("Metadata_Plate"),
                       operation = "mean",
                       na.rm = TRUE) %>%
  pivot_longer(-Metadata_Plate) %>%
  na.omit()

x2016_04_01_a549_48hr_batch1_SQ00015233_normalized_median_full_plate <-
  x2016_04_01_a549_48hr_batch1_SQ00015233_normalized %>%
  cytominer::aggregate(variables,
                       c("Metadata_Plate"),
                       operation = "median",
                       na.rm = TRUE) %>%
  pivot_longer(-Metadata_Plate) %>%
  na.omit()

x2016_04_01_a549_48hr_batch1_SQ00015233_normalized_mean_dmso <-
  x2016_04_01_a549_48hr_batch1_SQ00015233_normalized %>%
  filter(Metadata_broad_sample_type == "control") %>%
  cytominer::aggregate(variables,
                       c("Metadata_Plate"),
                       operation = "mean",
                       na.rm = TRUE) %>%
  pivot_longer(-Metadata_Plate) %>%
  na.omit()

x2016_04_01_a549_48hr_batch1_SQ00015233_normalized_median_dmso <-
  x2016_04_01_a549_48hr_batch1_SQ00015233_normalized %>%
  filter(Metadata_broad_sample_type == "control") %>%
  cytominer::aggregate(variables,
                       c("Metadata_Plate"),
                       operation = "median",
                       na.rm = TRUE) %>%
  pivot_longer(-Metadata_Plate) %>%
  na.omit()

x2016_04_01_a549_48hr_batch1_SQ00015233_normalized_mean_full_plate %>%
  ggplot(aes(value)) + 
  geom_histogram() +
  ggtitle("x2016_04_01_a549_48hr_batch1_SQ00015233_normalized_mean_full_plate")

x2016_04_01_a549_48hr_batch1_SQ00015233_normalized_median_full_plate %>%
  ggplot(aes(value)) + 
  geom_histogram() +
  ggtitle("x2016_04_01_a549_48hr_batch1_SQ00015233_normalized_median_full_plate")

x2016_04_01_a549_48hr_batch1_SQ00015233_normalized_mean_dmso %>%
  ggplot(aes(value)) + 
  geom_histogram() +
  ggtitle("x2016_04_01_a549_48hr_batch1_SQ00015233_normalized_mean_dmso")

x2016_04_01_a549_48hr_batch1_SQ00015233_normalized_median_dmso %>%
  ggplot(aes(value)) + 
  geom_histogram() +
  ggtitle("x2016_04_01_a549_48hr_batch1_SQ00015233_normalized_median_dmso")