distribution_mean.Rmd

---
title: Test File
---

```{r}
# Load necessary libraries
library(readr)
library(ggplot2)
```

```{r}
taxonomy <- 4
file_name_vector <- c("./subset_tables/subset_table_phylum.tsv",
                      "./subset_tables/subset_table_class.tsv",
                      "./subset_tables/subset_table_order.tsv",
                      "./subset_tables/subset_table_family.tsv",
                      "./subset_tables/subset_table_genus.tsv",
                      "./subset_tables/subset_table_species.tsv")
file_name <- file_name_vector[taxonomy]
```

```{r}
# Read the TSV file into R, skip the first row since it contains a description
mydata <- read_tsv(file_name,
                   skip = 1)

# Calculate mean and variance for each row
row_means <- apply(mydata[, -1], 1, mean)  # Assuming first column is row names
row_variances <- apply(mydata[, -1], 1, var)

```
```{r}
# Create a data frame to store the means and variances
mean_variance_data <- data.frame(
  Bacteria = mydata[[1]], # Assuming the first column contains bacteria names
  Mean = row_means,
  Variance = row_variances
)
```

```{r}
# Define the cutoff value for the mean
cutoff <- 0.001

# Filter the data to include only rows with Mean >= cutoff
mean_variance_data_filtered <- subset(mean_variance_data, Mean >= cutoff)

# Optional: Display the number of rows before and after filtering
total_rows <- nrow(mean_variance_data)
filtered_rows <- nrow(mean_variance_data_filtered)
cat("Total rows before filtering:", total_rows, "\n")
cat("Total rows after filtering:", filtered_rows, "\n")
cat("Number of rows removed:", total_rows - filtered_rows, "\n")
```

```{r}
# Plot the distribution of means
ggplot(mean_variance_data_filtered, aes(x = Mean)) +
  geom_histogram(binwidth = 0.005, fill = "blue", alpha = 0.7) +
  theme_minimal() +
  labs(title = "Distribution of Means", x = "Mean", y = "Frequency")
```

```{r}
# Plot Variance against Mean on Logarithmic X-axis
p <- ggplot(mean_variance_data_filtered, aes(x = Mean, y = Variance)) +
  geom_point(alpha = 0.6, color = "darkgreen") +
  theme_minimal() +
  labs(title = "Variance vs. Mean", x = "Mean (log scale)", y = "Variance") +
  scale_x_log10() +
  geom_smooth(method = "lm", se = FALSE, color = "red")  # Add a trend line
```

```{r}
# calculate the covariance of the mean_variance_data_filtered
ggsave("variance_mean.pdf",
       plot = p,
       width = 15,
       height = 8)
```


copy datasets to scc
scp -r /home/hongrui_liu/CAMH/American_Gut_R/subset_tables hliu@scclogin.camhres.ca:/nethome/kcni/hliu/projects/hliurepo