---
title: "Southen Ocean Diet and Energetics Data"
author: "Ben Raymond"
date: "`r Sys.Date()`"
output: rmarkdown::html_vignette
vignette: >
  %\VignetteIndexEntry{sohungry}
  %\VignetteEngine{knitr::rmarkdown}
  %\VignetteEncoding{UTF-8}
---

```{r chunkopts, eval = TRUE, echo = FALSE, warning = FALSE, message = FALSE}
knitr::opts_chunk$set(eval = TRUE, echo = TRUE, message = FALSE, warning = FALSE, tidy = FALSE, cache = FALSE, include = TRUE, dpi = 72, fig.width = 9, fig.height = 6, fig.align = "center", results = "markup")
library(knitr)
options(knitr.kable.NA = "")

## tidy up column names a bit
prettier <- function(z, first_to_capital = TRUE) {
  z <- gsub("_", " ", z)
  if (first_to_capital) z <- paste0(toupper(substr(z, 1, 1)), substr(z, 2, nchar(z)))
  z
}
```

## Overview

This R package provides access to the SCAR Southern Ocean Diet and Energetics Database, and some tools for working with these data. For more information about the database see http://data.aad.gov.au/trophic/.


### Installing

```{r eval = FALSE}
install.packages("devtools")
library(devtools)
install_github("SCAR/sohungry")
```

## Usage

Basic usage: load the desired dataset using `so_isotopes()`, `so_energetics()`, `so_lipids()`, `so_dna_diet()`, or `so_diet()`.

## Examples

```{r}
library(sohungry)
library(dplyr)
library(tidyr)
library(ggplot2)
```

### Isotopes

Load the stable isotope data, in measurement-value format (one row per measurement):
```{r}
xi <- so_isotopes(format = "mv")
```

Filter to taxon of interest, selecting d13C and d15N records:
```{r}
xi %>% dplyr::filter(taxon_name == "Electrona carlsbergi" & measurement_name %in% c("delta_13C", "delta_15N"))
```

### Diet

Load the diet data (stomach content analyses and similar):
```{r}
x <- so_diet()
```

A summary of what *Electrona carlsbergi* eats:
```{r eval = FALSE}
x %>% filter_by_predator_name("Electrona carlsbergi") %>% diet_summary(summary_type = "prey")
```
```{r echo = FALSE}
temp <- x %>% filter_by_predator_name("Electrona carlsbergi") %>% diet_summary(summary_type = "prey")
kable(temp, col.names = prettier(names(temp)), digits = 2)
```

And what eats *Electrona carlsbergi*:
```{r eval = FALSE}
x %>% filter_by_prey_name("Electrona carlsbergi") %>% diet_summary(summary_type = "predators")
```
```{r echo = FALSE}
temp <- x %>% filter_by_prey_name("Electrona carlsbergi") %>% diet_summary(summary_type = "predators")
kable(temp, col.names = prettier(names(temp)), digits = 2)
```

### Energetics

```{r}
xe <- so_energetics()
```

Select all single-individual records of *Electrona antarctica*:

```{r}
edx <- xe %>% dplyr::filter(taxon_sample_count == 1 & taxon_name == "Electrona antarctica")

## discard the dry-weight energy density values
edx <- edx %>% dplyr::filter(measurement_units != "kJ/gDW")

## some data manipulation
edx <- edx %>%
  ## remove the spaces from the measurement names, for convenience
  mutate(measurement_name = gsub("[[:space:]]+", "_", measurement_name)) %>%
  ## convert to wide format
  dplyr::select(source_id, taxon_sample_id, measurement_name, measurement_mean_value) %>%
  tidyr::spread(measurement_name, measurement_mean_value)

## what does this look like?
edx
```

Plot the wet weight against wet-weight energy density:

```{r ea_plot1}
p <- ggplot(edx, aes(wet_weight, energy_content))+geom_point()+theme_bw()+
  labs(x = "Wet weight (g)", y = "Energy density (kJ/g wet weight)")
plot(p)
```

Fit an allometric equation:

```{r ea_plot2}
fit <- lm(log(energy_content)~log(wet_weight), data = edx)
px <- tibble(wet_weight = seq(from = min(edx$wet_weight), to = max(edx$wet_weight), length.out = 51))
px$energy_content <- exp(predict(fit, newdata = px))
p+geom_path(data = px, colour = "dodgerblue")
```


### Lipids and fatty acids

```{r}
xl <- so_lipids()
```

Select lipid-class data from Connan et al. (2007), and plot similar to Figure 2 from that paper:

```{r lip_plot1}
xl <- xl %>% dplyr::filter(source_id == 126 & measurement_class == "lipid class") %>%
  mutate(measurement_name = sub(" content", "", measurement_name)) ## tidy the names a little

ggplot(xl,
  aes(measurement_name, measurement_mean_value, fill = taxon_life_stage, group = taxon_life_stage))+
  geom_col(position = "dodge")+theme_bw()+
  labs(x = "Lipid class", y = "Percentage of lipids")
```