JSRe0151-P_ChatGPT_qPCR-Processor.Rmd

---
title: "ChatGPT-qPCR-Processor"
author: "Jacob Roth"
date: "2024-02-24"
output: html_document
---

```{r setup, include=FALSE}
knitr::opts_chunk$set(echo = TRUE)
```

```{r}
# results <- "qPCR_20240201_1/20240208_1515_JSRe0151_151303_Results_20240208 165817.csv"
# overview <- "JSRe0151-P_DNA_qPCR-TidyqPCR_20240208_jsr.xlsx"
# exp <- "JSRe0151_20240201"

results <- "qPCR_20240223_1/20230223_JSRe0151_154845_20240223_181819_Results_20240223 200347.csv"
overview <- "JSRe0151-P_DNA_qPCR-TidyqPCR_jsr_20240212-2.xlsx"
exp <- "JSRe0151_20240208"

```

```{r}
# Load necessary libraries
library(dplyr)

# Read the data
data_qPCR <- read.csv("2-data_raw/qPCR_20240223_1/20230223_JSRe0151_154845_20240223_181819_Results_20240223 200347.csv",
                     skip = 21)

# data_qPCR <- read.csv(paste("2-data_raw/",results, sep = ""),
                     # skip = 21)
data_qPCR <- data_qPCR %>%
  rename(target = Target)  # Replace 'gene_target' with the actual column name
data_qPCR <- data_qPCR %>%
  rename(sample = Sample)  # Replace 'gene_target' with the actual column name

data_qPCR$sample_target <- paste(data_qPCR$sample, data_qPCR$target, sep="; ")

```

```{r remove primer dilutions}

data_qPCR <- data_qPCR %>%
  filter(!grepl("_200", sample)) %>%
  filter(!grepl("_2000", sample))

```


```{r Read metadata}
data_qPCR_primers <- read_excel(paste("1-notes/",overview, sep = ""), 
                              sheet = "PrimerEfficiency")
data_qPCR_primers <- dplyr::select(data_qPCR_primers, c(target,Efficiency_target)) %>% unique() %>%
  drop_na()



data_qPCR_overview <- read_excel(paste("1-notes/",overview, sep = ""), 
                              sheet = "overview")
data_qPCR_overview <- data_qPCR_overview %>%
  select(-c(sample,target))

# 
# gene_house <- "Actin_JSRi032-033"
# sample_control1 <- "JSRp033_PolyA_20"
# sample_control2 <- "DMSO-4day_PolyA_20"


```

```{r Align identifiers}

#add metadata for relevent controls
data_qPCR <- data_qPCR %>%
  left_join(data_qPCR_overview, by = "sample_target")


#add primer efficiency column for target
data_qPCR <- data_qPCR %>%
  left_join(data_qPCR_primers, by = "target")

#add primer efficiency column for control
data_qPCR <- merge(data_qPCR, data_qPCR_primers, by.x = "gene_house1", by.y = "target", all.x = TRUE)

data_qPCR <- data_qPCR %>%
  rename(Efficiency_target = Efficiency_target.x)

data_qPCR <- data_qPCR %>%
  rename(Efficiency_control = Efficiency_target.y)

```

```{r Processing}

# Calculate the control Cq for each sample and target
data_qPCR_control_cq <- data_qPCR %>%
  group_by(sample) %>%
  summarise(Cq_control = mean(as.numeric(Cq[gene_house1 == target], na.rm = FALSE)), .groups = 'drop')

# Join this control Cq back to the original dataset
data_qPCR <- data_qPCR %>%
  left_join(data_qPCR_control_cq, by = c("sample"))

```

```{r Calc deltaCq}
data_qPCR$Cq <- as.numeric(data_qPCR$Cq)
data_qPCR$Cq_control <- as.numeric(data_qPCR$Cq_control)
data_qPCR$Efficiency_target <- as.numeric(data_qPCR$Efficiency_target)
data_qPCR$Efficiency_control <- as.numeric(data_qPCR$Efficiency_control)

# data_qPCR$RelAbundance <- ((data_qPCR$Efficiency_target^(-data_qPCR$Cq))/(data_qPCR$Efficiency_control^(-data_qPCR$Cq_control)))

data_qPCR$RelAbundance <- ((data_qPCR$Efficiency_target^(-data_qPCR$Cq)) / (data_qPCR$Efficiency_control^(-data_qPCR$Cq_control)))


```
```{r}
data_qPCR$sample_target_control1 <- paste(data_qPCR$sample_control1,"; ",data_qPCR$target,sep = "")


# Step 1 & 2: Filter rows where sample_target_control1 equals sample_target and select Cq
matching_Cq_data <- data_qPCR %>%
  filter(sample_target_control1 == sample_target) %>%
  select(sample_target_control1, RelAbundance)  # Assuming you want to keep the sample and target columns

matching_Cq_data <- matching_Cq_data %>%
  rename(RelAbundance_control = RelAbundance)

data_qPCR <- data_qPCR %>%
  left_join(matching_Cq_data, by = "sample_target_control1")



data_qPCR$FC <- data_qPCR$RelAbundance/data_qPCR$RelAbundance_control
```


```{r Prism Save}
data_qPCR_PRISM <- data_qPCR %>%
  select(c(sample, target,FC)) %>%
  na.omit()
data_qPCR_PRISM <- pivot_wider(data_qPCR_PRISM,
                    names_from = sample, values_from = FC,
                    values_fill = 0)


# write.csv(summary_RI,
#           file='20221207_MergedRMATs-Encode-Public-Shechter_Summary-RI_FDR-filtered.csv',
#           row.names = FALSE
#           )
write.csv(data_qPCR_PRISM,
          file=paste("4-data_processed/",exp,'_DataForPrism.csv', sep =""),
          row.names = FALSE)


```


```{r, eval = FALSE}
library(dplyr)

# Calculate average Cq per Sample, assuming Cq is already numeric
# data_summary <- data %>%
#   group_by(Sample) %>%
#   summarise(Avg_Cq = mean(Cq, na.rm = TRUE))

library(ggplot2)

# Plotting
ggplot(data_filtered, aes(x = Sample, y = Relative_Difference, fill = Sample)) +
  geom_bar(stat = "identity", position = "dodge") +
  theme_minimal() +
  labs(title = "Average Cq Values by Sample",
       x = "Sample",
       y = "Average Cq",
       fill = "Sample") +
  theme(axis.text.x = element_text(angle = 45, hjust = 1)) # Rotate x-axis labels for readability
  # facet_grid()


```