add beginning of GI review

R/05-crispr-calc.R

@@ -175,9 +175,7 @@ calc_crispr <- function(.data = NULL,
       pgRNA2_seq = gRNA2_seq,
       mean_double_control_crispr_1 = mean_double_control_crispr,
       mean_double_control_crispr_2
-    ) %>%
-    # TODO: What we dropping here? This needs investigation
-    dplyr::distinct()
+    )
 
   # Save at the target level
   gimap_dataset$crispr_score <- crispr_df

R/06-calculate_gi.R

@@ -99,7 +99,8 @@ calc_gi <- function(.data = NULL,
       rep,
       double_target_gi_score,
       single_target_gi_score_1,
-      single_target_gi_score_2
+      single_target_gi_score_2,
+      expected_crispr
     )
 
   # Store this

inst/rmd/scratch_gimap_GI_review.Rmd

@@ -0,0 +1,87 @@
+---
+title: "scratch gimap Genetic Interaction Review"
+output: html_document
+date: "`r Sys.Date()`"
+---
+
+```{r warning=FALSE, message=FALSE}
+library(tidyverse)
+library(devtools)
+
+devtools::load_all()
+```
+
+## Get the gimap GI results
+
+```{r}
+ gimap_dataset <- get_example_data("gimap") %>%
+   gimap_filter() %>%
+   gimap_annotate() %>%
+   gimap_normalize(
+     timepoints = "day",
+     replicates = "rep") %>%
+  calc_crispr() %>%
+  calc_gi()
+```
+
+```{r}
+saveRDS(gimap_dataset, "gimap_dataset_calculated.RDS")
+```
+
+## Load the GI Mapping GI results
+
+```{r}
+old_gi_results <- readRDS("d.HeLa_GI_scores_target")
+```
+
+## Compare results
+
+### Columns
+
+```{r}
+colnames(gimap_dataset$gi_scores)
+```
+
+```{r}
+colnames(old_gi_results)
+```
+
+Looks like we want to compare the pgRNA targets which should be the `pgRNA_target_double` column for the gimap results and the `pgRNA_target` column for the GI Mapping results
+
+Columns in the GI Mapping results not in the gimap results seem to include model results (`intercept`, `slope`) as well as some significance values (`p_val`, `fdr`), observed and expected CRISPR score (`mean_observed_CS` and `mean_expected_CS` (which I think I need for plotting)). And how does `mean_GI_score` relate to the three GI score columns in the gimap results? 
+
+### Target overlap
+
+```{r}
+
+length(unique(gimap_dataset$normalized_log_fc$pgRNA_target))
+
+length(unique(old_gi_results$pgRNA_target))
+length(unique(gimap_dataset$gi_scores$pgRNA_target_double))
+length(intersect(unique(old_gi_results$pgRNA_target), unique(gimap_dataset$gi_scores$pgRNA_target_double)))
+```
+
+Only 1/3 of the GI Mapping targets are represented in the gimap results
+
+### Number of observations/rows total and rep column overlap
+
+```{r}
+nrow(old_gi_results)
+nrow(gimap_dataset$gi_scores)
+```
+
+The gimap results has far more rows 
+
+```{r}
+table(old_gi_results$rep)
+
+table(gimap_dataset$gi_scores$rep)
+```
+
+While the replicate values themselves can be synced like we've done in other reviews, this further shows the far greater number of observations (even if there are fewer represented pgRNA targets) in the gimap results
+
+```{r}
+head(old_gi_results)
+
+```
+