---
title: "How to find the best fitting distribution"
output: rmarkdown::html_vignette
vignette: >
  %\VignetteIndexEntry{How to find the best fitting distribution}
  %\VignetteEngine{knitr::rmarkdown}
  %\VignetteEncoding{UTF-8}
---

```{r, include = FALSE}
knitr::opts_chunk$set(
  collapse = TRUE,
  comment = "#>"
)
```

```{r setup}
library(peppwR)
```




The package has one function for estimating the fit of a range of distributions `find_fits()`.  It takes a data frame with at least three columns:

  * `id_col` a unique id for the phosphopeptide 
  * `group_col` a category that groups measurements from the same bio replicate together
  * `value_col` the measurment abundance/counts etc


```{r, eval=FALSE}
df_path <- system.file("raw_data.csv", package="peppwr")
df <- readr::read_csv(df_path)

fits_df <- find_fits(df, id_col="pep_id", group_col="group", value_col="value")
```


It returns a nested dataframe with elements for AIC and LogLikelihood of fit of each peptide's counts to a range of distributions.


The result can be plot with `evaldist()`


```{r, eval=FALSE}
evaldist(fits_df)
```