taxonomy_distribution.Rmd

---
title: "taxonomy_distribution"
output: html_document
date: "2024-11-13"
---

```{r}
library(readr)
library(ggplot2)
library(tidyverse)

avg_dist_df <- read_csv("workflow/out/taxonomy/sample_1_0_02_30Lw30Nw1.0p1.0q150k_species_averageDistance.csv")
dist_species_df <- read_csv("workflow/out/taxonomy/sample_1_0_02_30Lw30Nw1.0p1.0q150k_species_distancesWithTax.csv", guess_max = 10000)
species_df <- read_csv("workflow/out/node_classification/sample_1_0_02_nodes_by_species.csv", guess_max = 100000)
degree_df <- read_csv("workflow/out/node_classification/sample_1_0_02_node_degrees.csv", guess_max = 100000)
```

# inter vs intra species avg pairwise euclidean distances
```{r}
ggplot(subset(avg_dist_df, species != "multiple_species"), aes(x = average_pw_euclidean_distance_same_species, average_pw_euclidean_distance_cross_species)) +
  geom_point() +
  xlab(paste("Avg pairwise euclidean distance", "for same-species node pairs", sep="\n")) +
  ylab(paste("Avg pairwise euclidean distance", "for cross-species node pairs", sep="\n")) +
  ggtitle("Same- vs. cross-species average pairwise euclidean distances")


avg_dist_df_long <- avg_dist_df
  pivot_longer(!species, names_to = "species_grouping", values_to = "avg_pairwise_euclidean_distance")

avg_dist_df_long$species_grouping[avg_dist_df_long$species_grouping == "average_pw_euclidean_distance_same_species"] <- "same_species"

avg_dist_df_long$species_grouping[avg_dist_df_long$species_grouping == "average_pw_euclidean_distance_cross_species"] <- "cross_species"
  
ggplot(subset(avg_dist_df_long, species != "multiple_species"), aes(x = species_grouping, avg_pairwise_euclidean_distance)) +
  geom_boxplot() 
  xlab(paste("Avg pairwise euclidean distance", "for same-species node pairs", sep="\n")) +
  ylab(paste("Avg pairwise euclidean distance", "for cross-species node pairs", sep="\n")) +
  ggtitle("Same- vs. cross-species average pairwise euclidean distances")

```


```{r}
dist_species_df_classes <- dist_species_df %>%
  mutate(same_species = case_when(species_i == species_j ~ "same_species", 
                                species_i != species_j ~ "different_species"), 
         pair_species = case_when(species_i == species_j ~ species_i, 
                                species_i != species_j ~ NA)) 


```

```{r}
ggplot(subset(dist_species_df_classes, species_i != "multiple_species" | species_j != "multiple_species"), aes(x = distance, color = same_species)) +
  geom_density()


```

```{r}
species_deg_nodes <- species_df %>%
  
```