WhoseEgg

.title[
# WhoseEgg
]
.subtitle[
## .white[A Shiny App for Identifying Invasive Carp Using Random Forests and Fish Egg Characteristics]
]
.author[
### <br>
]
.author[
### .white[Katherine Goode]
]
.author[
### .white[Joint work with] <a href="https://michaeljohnweber.wixsite.com/fisherieslab">Dr. Michael Weber</a> .white[and] <a href="https://pdixon.stat.iastate.edu/">Dr. Philip Dixon</a>
]
.author[
### .white[Presented at ISU LunchinatoRs]
]
.date[
### .white[April 16, 2021] <br> <br> <br> <br> .small[.white[Code used to generate slides found <a href="https://github.com/goodekat/presentations/tree/master/2021-lunchinatoRs-shiny-fish">here</a>]]
]

---

# Overview

- Background on R Shiny
  
- Project Background

- WhoseEgg App

<br>

<center> ![](figures/jumping-carp.gif) <center>

---

# Background on R Shiny

---

# What is Shiny?

Shiny lets you create apps using R!

From the [Shiny website](https://shiny.rstudio.com/):

> Shiny is an R package that **makes it easy to build interactive web apps straight from R**. You can host standalone apps on a webpage or embed them in R Markdown documents or build dashboards. You can also extend your Shiny apps with CSS themes, htmlwidgets, and JavaScript actions.

---

# Shiny website

<body>
<style type="text/css">
body, html
{
margin: 0; padding: 0; height: 100%; overflow: hidden;
}
#content
{
position:absolute; left: 0; right: 0; bottom: 0; top: 0%; 
}
</style>
<div id="content">
<iframe width="100%" height="100%" frameborder="0" src="https://shiny.rstudio.com/"/>
</div>
</body>

---

# Project Background

---

# Invasive Carp

- Brought to US in 1960s

- Become a threat to native fish populations in Mississippi River basin

- Interest in monitoring population to prevent further spread

<br>

---

# Egg Collection

- One method for monitoring carp is to collect eggs

- Common practice is to have eggs identified using genetic analysis

- Expensive

- Led [Camacho et al. (2019)](https://afspubs.onlinelibrary.wiley.com/doi/abs/10.1002/nafm.10380) to try using machine learning to classify fish eggs

<br>

.center[
<img src="figures/egg1.png" width="30%" /><img src="figures/egg2.png" width="30%" /><img src="figures/egg5.png" width="30%" />
]

---

# Machine Learning Appraoch

.pull-left[
**[Camacho et al. (2019)](https://afspubs.onlinelibrary.wiley.com/doi/abs/10.1002/nafm.10380)**

- Trained random forests

- Predictor variables: 
  - Egg characteristics

- Responses: 
  - Family, genus, and species

- Models showed promise for identifying invasive carp

]

**[Goode et al. (2021)](https://afspubs.onlinelibrary.wiley.com/doi/10.1002/nafm.10616)**

- Validated models using a new year of data

- Trained new models with all years of data

]

---

# Sampling Site and Years

Three years of collection

---

# Egg Collection

Collected eggs measured and genetically identified

---

# Predictor Variables

- Month
- Julian day  
- Water temperature	 
- Water conductivity  	
- Deflated?	 
- Pigment?  
- Compact or diffuse?  	
- Sticky debris?  
- Egg stage  
- Membrane average  
- Membrane standard deviation  
- Membrane coefficient of variation  
- Embryo average
- Embryo standard deviation  
- Embryo coefficient of variation
- Embryo to membrane ratio
- Larval length
]

<img src="figures/egg-characteristics.png" width="100%" style="display: block; margin: auto;" />
]

---

# Response Variables

**Taxonomy with invasive carp categorized as one class**:

- Family (8 levels)

- Genus (16 levels)

- Species (27 levels)

]

<br> 
<small> Image source: https://www.watershedcouncil.org/asian-carp.html </small>
]

---

# Validation of Original Models

Metrics computed using 2014-2015 model on 2016 data

---

# WhoseEgg App

---

# WhoseEgg

**Goals**

- Provide a tool that is accessible to non-R users for using the random forest to predicting the identification of fish eggs

- Used to help identify areas where invasive carp are reproducing

- Screen eggs to select a subset to send for genetic identification

<br>

**Models used**

- Random forests:

- Trained with all three years of data
  
  - Responses of family, genus, and species with invasive carp grouped

---

# App Demonstration

App available at: https://whoseegg.stat.iastate.edu/

Example datasets available at: https://github.com/goodekat/presentations/blob/master/2021-lunchinatoRs-shiny-fish/example_data.zip

---

# Possible Updates

<br>

- Add visualizations of input data compared to training data

<br>

- Include prediction intervals

<br>

- Add visualizations relating to explainability for individual predictions

<br>

- Add video showing how to use the app

<br>

- Add image during loading of swimming fish

---

# Thank you!

---

# Additional Figures

---

# Metrics on Training Data

Models trained with 2014-2015 and 2014-2016 data

---

# Metrics on New/Old Sites

---

# Feature Importance

---

# Partial Dependence Plots

---

# Correlation Between Features