---
title: "Getting started with epifitter"
author: "Kaique S. Alves"
date: "`r Sys.Date()`"
output: rmarkdown::html_vignette
vignette: >
  %\VignetteIndexEntry{Getting started with epifitter}
  %\VignetteEncoding{UTF-8}
  %\VignetteEngine{knitr::rmarkdown}
editor_options:
  chunk_output_type: inline
---

```{r setup, message=FALSE, warning=FALSE}
library(epifitter)
library(dplyr)
library(ggplot2)
library(cowplot)
theme_set(cowplot::theme_half_open(font_size = 12))
```

## What epifitter does

`epifitter` helps analyze plant disease progress curves by combining:

- simulation helpers for synthetic epidemics;
- fitting functions for canonical disease progress models;
- summary measures such as `AUDPC()` and `AUDPS()`;
- plotting functions that return `ggplot2` objects.

## A first epidemic

```{r}
set.seed(1)

epi <- sim_logistic(
  N = 80,
  y0 = 0.01,
  dt = 10,
  r = 0.12,
  alpha = 0.2,
  n = 5
)

knitr::kable(head(epi), digits = 4)
```

```{r fig.alt="Plot of a simulated epidemic showing replicate observations and the underlying disease progress curve over time."}
ggplot(epi, aes(time, y, group = replicates)) +
  geom_point(aes(y = random_y), shape = 1, color = "#8597a4") +
  geom_line(color = "#15616d", linewidth = 0.8) +
  labs(
    title = "Simulated epidemic",
    x = "Time",
    y = "Disease intensity"
  )
```

## Fit candidate models

```{r}
fit <- fit_lin(time = epi$time, y = epi$random_y)
fit
```

`fit$stats_all` contains the full ranking of candidate models.

```{r}
knitr::kable(fit$stats_all, digits = 4)
```

## Visualize predictions

```{r fig.alt="Faceted plot comparing observed disease intensity values with fitted curves from candidate models."}
plot_fit(fit, point_size = 1.8, line_size = 0.9)
```

## Work with multiple epidemics

```{r}
epi_a <- sim_gompertz(N = 50, y0 = 0.002, dt = 5, r = 0.10, alpha = 0.2, n = 3)
epi_b <- sim_gompertz(N = 50, y0 = 0.002, dt = 5, r = 0.14, alpha = 0.2, n = 3)

multi_epi <- bind_rows(epi_a, epi_b, .id = "epidemic")

multi_fit <- fit_multi(
  time_col = "time",
  intensity_col = "random_y",
  data = multi_epi,
  strata_cols = "epidemic"
)

knitr::kable(head(multi_fit$Parameters), digits = 4)
```

## Next steps

- Use the model fitting article for a deeper walkthrough of `fit_lin()`, `fit_nlin()`, `fit_nlin2()`, and `fit_multi()`.
- Use the simulation article for examples built around the `sim_` family.
- Use the PowderyMildew article for a real-data workflow based on the bundled experimental dataset.