talk06 - part 1.Rmd

---
title: "R for bioinformatics, data wrangler, part 2"
subtitle: "HUST Bioinformatics course series"
author: "Wei-Hua Chen (CC BY-NC 4.0)"
institute: "HUST, China"
date: "`r format(Sys.time(), '%d %B, %Y')`"
output: 
  beamer_presentation:
    theme: AnnArbor
    colortheme: beaver
    fonttheme: structurebold
    highlight: tango
    includes:
      in_header: mystyle.sty
---

```{r include=FALSE}
color_block = function(color) {
  function(x, options) sprintf('\\color{%s}\\begin{verbatim}%s\\end{verbatim}',
                               color, x)
}

## 将错误信息用红色字体显示
knitr::knit_hooks$set(error = color_block('red'))
```

# section 1: TOC

## 前情提要

### pipe
-   pipe

### dplyr

-   select()
-   filter()
-   mutate()
-   summarise()
-   arrange()
-   group_by() ...

## 今次提要

### tidyr

-   1. 长宽数据转变
-   2. 数据分割：一列变多列
-   3. 数据合并：多列变一列
-   4. 其它函数

# section 2: data wrangler - tidyr

## tidyr

### what is `tidyr` ?

The goal of tidyr is to help you create **tidy** data.

![dplyr logo](images/talk06/tidyr.png){height="30%"}


### more to read

-   [tidyr offical page at tidyverse](https://tidyr.tidyverse.org)
-   [R for data science](https://r4ds.had.co.nz)

## tidyr 安装

只需**安装一次**即可！

\FontSmall

```{r eval=FALSE}
# The easiest way to get tidyr is to install the whole tidyverse:
install.packages("tidyverse")

# Alternatively, install just tidyr:
install.packages("tidyr")

# Or the development version from GitHub:
# install.packages("devtools")
devtools::install_github("tidyverse/tidyr")
```

\FontNormal

[Get the cheatsheet at here](https://github.com/rstudio/cheatsheets/blob/master/tidyr.pdf)

## 1. 长宽数据转变：`宽数据` 向 `长数据` 转变

get data ready 

\FontSmall

```{r message=FALSE, warning=FALSE}
library(tidyverse); ## 先装入包；
grades2 <- read_tsv(file = "data/talk06/grades2.txt");

grades2;
```

## 宽数据的特点

### 优点：

-    自然，易理解；

### 缺点：

-    不易处理；
-    稀疏时问题较大；

## `宽数据` 向 `长数据` 转变

\FontSmall

```{r message=F}
library(kableExtra);
grades3 <- grades2 %>% pivot_longer( - name, names_to = "course", values_to = "grade" );
kbl( grades3 );
```

## `pivot_longer` explained!

\FontSmall

```{r eval=FALSE}
grades3 <- grades2 %>% pivot_longer( - name, names_to = "course", values_to = "grade" );
```

\FontNormal

![ `pivot_longer` explained! ](images/talk05/gather_explained.png){height="50%"}

## 有 `NA` 值怎么办？

\FontSmall

```{r}
grades3_1 <- grades3[ !is.na(grades3$grade),  ];
grades3_2 <- grades3[ complete.cases( grades3 )  ,  ];

## -- 更好的方法 ~~ 
grades3_long <- grades2 %>% 
  pivot_longer( - name, 
                names_to = "course", 
                values_to = "grade",
                values_drop_na = TRUE);
```


`values_drop_na` 即可消除；


## 有 `NA` 值怎么办？cont.

\FontSmall

```{r}
kbl( grades3_long );
```

## 长变宽

\FontSmall

```{r}
grades3_wide <- grades3_long %>% 
  pivot_wider( names_from = "course", values_from = "grade" );

grades3_wide;
```

## `pivot_wider` 怎么用？

![`pivot_wider` function explained](images/talk05/wider_explained.png){height="50%"}

## 宽长数据转换练习

用 `pivot_wider` 和 `pivot_longer` 对下面的数据 `mini_iris` 进行宽长转换:

\FontSmall

```{r}
mini_iris <- iris[ c(1, 51, 101),  ];
kbl( mini_iris);
```

\FontNormal

`iris` 是 鸢尾属 一些物种花瓣的量表

## 宽变长， cont.

\FontSmall

```{r}
## -- 注意： 第一、二个参数可以自行命名，分别对应原始数据中的 column names 及 values ...
mini_iris.longer <- mini_iris %>% 
  pivot_longer( - Species, names_to = "type", values_to = "dat" );
kbl( mini_iris.longer  );
```


## 长变宽

\FontSmall

```{r}
## -- 注意： 第一、二个参数可以自行命名，分别对应原始数据中的 column names 及 values ...
mini_iris.wider <- mini_iris.longer %>% 
  pivot_wider( names_from =  "type", values_from = "dat" );
kbl( mini_iris.wider  );
```

## 比较复杂的例子

\FontSmall

```{r message=FALSE, warning=FALSE}
grades2 <- read_delim( file = "data/talk05/grades2.txt", delim = "\t",
                       quote = "", col_names = T);
kbl( grades2 );
```

\FontNormal

这是哪种数据类型？长还是宽？？

## 怎么变成宽数据？

\FontSmall

```{r}
grades2_wide <- grades2 %>% 
  pivot_wider( names_from = course, values_from =  grade );

grades2_wide;
```

## 再变成长数据

又怎么把它变回来？

\FontSmall

```{r}
a <- 
  grades2_wide %>% 
  pivot_longer( ! c( name, class ), 
                names_to = "course", 
                values_to = "grade", 
                values_drop_na = T 
               );

kbl( a );
```

## 另一种变法, 注意两者的区别！！

\FontSmall

```{r}
b <- grades2_wide %>% 
  pivot_longer( bioinformatics:spanish, ## 选择成绩所在的列！
                names_to = "course", values_to = "grade", 
                               values_drop_na = T 
                               );

kbl( b );
```

## 2. tidyr::separate 将`一列`拆成`多列`

\FontSmall

```{r}
table3 <- read_tsv(file = "data/talk06/table3.txt");

table3 %>% 
  separate(rate, into = c("cases", "population"), sep = "/");

```

## tidyr::separate 同时进行格式转换

如何把分拆后的列`正确识别为数字`？？

\FontSmall

```{r}
table3 %>% 
  separate(rate, into = c("cases", "population"), convert = TRUE)
```

## tidyr::separate 按字符长度分割

把年拆分为`世纪`和`年`

\FontSmall

```{r}
table5 <- table3 %>% 
  separate(year, into = c("century", "year"), sep = 2)

table5;
```

## 3. tidyr::unite 将`多列`合成`一列`

\FontSmall

```{r}
## 将上页分拆的结果进行合并

table5 %>% 
  unite(new, century, year, sep = ""); ## sep 参数默认是 _ 

```

## `seperate` 与 `unit` 小结

-    分割和合并后，原列会`消失`! 可使用 `remove = FALSE` 保留原列
-    更多示例见：<https://r4ds.had.co.nz/tidy-data.html>

## 4. 其它函数

-    `fill` （作业会用到）
-    `complete` （作业会用到）
-    ...

更多示例见：<https://r4ds.had.co.nz/tidy-data.html>


# section 3 : 小结与作业

## 小结

### 今次提要

-   tidyr (超级强大的数据处理) part 2

### 下次预告

-   dplyr, tidyr 和 forcats 的更多功能与生信操作实例

### important

-   all codes are available at Github: <https://github.com/evolgeniusteam/R-for-bioinformatics>

## 练习 & 作业

-   `Exercises and homework` 目录下 `talk06-homework.Rmd` 文件；

-   完成时间：见钉群的要求