---
title: "Partially Matching of Trait Data and Tree(s) in treedata.table"
author: "Josef Uyeda, Cristian Roman-Palacios, April Wright"
date: "08/08/2020"
output: rmarkdown::html_vignette
vignette: >
  %\VignetteIndexEntry{Partially Matching of Trait Data and Tree(s) in treedata.table}
  %\VignetteEngine{knitr::rmarkdown}
  \usepackage[utf8]{inputenc}
---

```{r setup, include=FALSE}
knitr::opts_chunk$set(echo = TRUE)
```

## Partially matching trait data and tree(s)

The `as.treedata.table` function enables users to match a tree (or multiple trees) against a single trait database. We first load the sample dataset.

```{r}
library(ape)
library(treedata.table)

# Load example data
data(anolis)
#Create treedata.table object with as.treedata.table
td <- as.treedata.table(tree = anolis$phy, data = anolis$dat)
```


Tips that are not common between the tree (or trees) and dataset are dropped from the resulting `treedata.table` object. For instance, below I have modified the original anole phylogeny such that *A. ahli* (**ahli**) is replaced for a label that is not present in the dataset (**NAA**).

```{r}
anolis_newtip<-anolis$phy
anolis_newtip$tip.label[1]<-'NAA'
anolis_newtip
```


We then use this modified tree to fit a `treedata.table` object using the `as.treedata.table` function:


```{r}
td <- as.treedata.table(tree=anolis_newtip, data=anolis$dat)
```

Note that `as.treedata.table` drops all non-overlapping tips (**NAA** [present in the tree but not in the trait data] and **ahi** [present in the database but not in tree] in this case) and returns a `treedata.table` object with fully matching `phy` and `data` objects. 

```{r}
td
```


Fully-matching matrix and trees are also returned in `treedata.table` objects with `multiPhylo` objects in their `phy` component. See the example below.

We first construct a `multiPhylo` object that partially overlaps the original trait database by using **NAA** instead of **ahi**.

```{r}
anolis2<-anolis$phy
anolis2$tip.label[1]<-'NAA'
anolis1<-anolis$phy
anolis1$tip.label[1]<-'NAA'
trees<-list(anolis1,anolis2)
class(trees) <- "multiPhylo"
trees
```

Next, we fit the `treedata.table` object using the relevant `multiPhylo` object and the original trait database.

```{r}
td <- as.treedata.table(tree=trees, data=anolis$dat)
```

Note that 1 tip was dropped for all trees in the `multiPhylo` object and a single row was deleted from the `data.table` object in the `treedata.table` object.

```{r}
td
```