An introduction to (problem solving with) R

class: center, middle, inverse, title-slide

# An introduction to (problem solving with) R
### Sina Rüeger
### 2018-09-12 (updated: 2018-09-13)

---

---
layout: true

---
class: left, middle

# About me

- Background in Data Analysis / Data Science

- PhD in Life Sciences @ CHUV PostDoc @ EPFL (analysis of genetic data)

- -Ladies Lausanne co-organiser

- Data analysis & Genetic data & Data visualisation

---
class: center, middle, inverse

# What is R?

---
class: left, middle

# R ...

- is "a programming language for statistical computing"

- is free

- has a webpage: https://www.r-project.org/

- just celebrated its [25 year anniversary](https://rss.onlinelibrary.wiley.com/doi/10.1111/j.1740-9713.2018.01169.x)

- comes with *basic*/*default* packages, but there are over 13'000 R-packages 😮 that can be installed through [CRAN](https://cran.r-project.org/web/packages/) or repositories like github

---
class: center, middle

# Typical data analysis workflow

## 🤔

`Question > Data > Analyse data with a tool > Distill knowlege from data > Feel enlightened :-) > Take decisions`

---
class: inverse, center, middle

# What is R used for?

---
class: middle

# Quite a few things

- In general lots of (but not all) data science, data analysis & stats stuff

- Me: genomics, biostatistics

- To make presentations (this one)

---

## Basic R

### Math operations

```r
1 + 1  ## basic math operations
```

```
## [1] 2
```

### Important constants

```r
pi
```

```
## [1] 3.141593
```

---

## Basic R

### Vector, matrices

```r
a <- c(34, 1, 67) ## defining a vector
mat <- matrix(1:6, ncol = 2) ## defining a matrix
class(a)
```

```
## [1] "numeric"
```

```r
class(mat)
```

```
## [1] "matrix"
```

---

## Basic R 
### Object oriented

```r
summary(a)
```

```
##    Min. 1st Qu.  Median    Mean 3rd Qu.    Max. 
##     1.0    17.5    34.0    34.0    50.5    67.0
```

```r
summary(mat)
```

```
##        V1            V2     
##  Min.   :1.0   Min.   :4.0  
##  1st Qu.:1.5   1st Qu.:4.5  
##  Median :2.0   Median :5.0  
##  Mean   :2.0   Mean   :5.0  
##  3rd Qu.:2.5   3rd Qu.:5.5  
##  Max.   :3.0   Max.   :6.0
```

---

## Basic R

### Asking for help

```r
`?`(data.frame)
```

### Making a plot

```r
plot(rnorm(100))  ## plot 100 numbers that were drawn randomly from a normal distribution
```

![](slides_files/figure-html/unnamed-chunk-6-1.png)

---

## Statistical analyses

### 1. Load libraries

```r
## install packages (only run this once) install.packages('readr')
## install.packages('dplyr') install.packages('skimr')
## install.packages('ggplot2')

library(readr)  ## for read_csv()
library(dplyr)  ## for rename()
library(skimr)  ## for skim()
library(ggplot2)  ## for visualisations
theme_set(theme_bw())
```
---

### 2. Import data

```r
dat <- read_csv("https://vincentarelbundock.github.io/Rdatasets/csv/datasets/swiss.csv")
# use library(readxl) for xls data! swiss is actually an R package dataset
# of library(datasets), but has the column X1 as rownames
`?`(swiss)
dat
```

```
## # A tibble: 47 x 7
## X1 Fertility Agriculture Examination Education Catholic
## <chr> <dbl> <dbl> <int> <int> <dbl>
## 1 Cour… 80.2 17 15 12 9.96
## 2 Dele… 83.1 45.1 6 9 84.8 
## 3 Fran… 92.5 39.7 5 5 93.4 
## 4 Mout… 85.8 36.5 12 7 33.8 
## 5 Neuv… 76.9 43.5 17 15 5.16
## 6 Porr… 76.1 35.3 9 7 90.6 
## 7 Broye 83.8 70.2 16 7 92.8 
## 8 Glane 92.4 67.8 14 8 97.2 
## 9 Gruy… 82.4 53.3 12 7 97.7 
## 10 Sari… 82.9 45.2 16 13 91.4 
## # ... with 37 more rows, and 1 more variable: Infant.Mortality <dbl>
```
---

### 3. Summarise data

```r
## summary
skim(dat %>% select(X1, Fertility, Education, Catholic))
```

```
## Skim summary statistics
##  n obs: 47 
##  n variables: 4 
## 
## ── Variable type:character ──────────────────────────────────────────────────────────────────────────────────────────────────────────────────────────────────
##  variable missing complete  n min max empty n_unique
##        X1       0       47 47   4  12     0       47
## 
## ── Variable type:integer ────────────────────────────────────────────────────────────────────────────────────────────────────────────────────────────────────
##   variable missing complete  n  mean   sd p0 p25 p50 p75 p100     hist
##  Education       0       47 47 10.98 9.62  1   6   8  12   53 ▇▇▁▁▂▁▁▁
## 
## ── Variable type:numeric ────────────────────────────────────────────────────────────────────────────────────────────────────────────────────────────────────
##   variable missing complete  n  mean    sd    p0  p25   p50   p75  p100
##   Catholic       0       47 47 41.14 41.7   2.15  5.2 15.14 93.12 100  
##  Fertility       0       47 47 70.14 12.49 35    64.7 70.4  78.45  92.5
##      hist
##  ▇▂▁▁▁▁▁▅
##  ▁▁▁▂▇▅▃▂
```
---

### 4. Rename a column

```r
## colum to rownames + binary catholic (> 50%)
dat <- dat %>% rename(Region = X1) %>% mutate(Catholic.bin = Catholic > 50)
dat
```

```
## # A tibble: 47 x 8
## Region Fertility Agriculture Examination Education Catholic
## <chr> <dbl> <dbl> <int> <int> <dbl>
## 1 Court… 80.2 17 15 12 9.96
## 2 Delem… 83.1 45.1 6 9 84.8 
## 3 Franc… 92.5 39.7 5 5 93.4 
## 4 Mouti… 85.8 36.5 12 7 33.8 
## 5 Neuve… 76.9 43.5 17 15 5.16
## 6 Porre… 76.1 35.3 9 7 90.6 
## 7 Broye 83.8 70.2 16 7 92.8 
## 8 Glane 92.4 67.8 14 8 97.2 
## 9 Gruye… 82.4 53.3 12 7 97.7 
## 10 Sarine 82.9 45.2 16 13 91.4 
## # ... with 37 more rows, and 2 more variables: Infant.Mortality <dbl>,
## # Catholic.bin <lgl>
```

---

### 5. Visualise data

```r
ggplot(data = dat) + geom_point(aes(Education, Fertility, color = Catholic))
```

![](slides_files/figure-html/data-vis-1.png)
---

### 7. Fit model

```r
mod <- lm(Fertility ~ Education + Catholic.bin, data = dat)
summary(mod)
```

```
## 
## Call:
## lm(formula = Fertility ~ Education + Catholic.bin, data = dat)
## 
## Residuals:
## Min 1Q Median 3Q Max 
## -17.739 -5.832 -1.953 6.251 15.466 
## 
## Coefficients:
## Estimate Std. Error t value Pr(>|t|) 
## (Intercept) 75.9378 2.3091 32.886 < 2e-16 ***
## Education -0.8006 0.1355 -5.909 4.59e-07 ***
## Catholic.binTRUE 7.8173 2.6512 2.949 0.0051 ** 
## ---
## Signif. codes: 0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
## 
## Residual standard error: 8.729 on 44 degrees of freedom
## Multiple R-squared: 0.5329,	Adjusted R-squared: 0.5117 
## F-statistic: 25.1 on 2 and 44 DF, p-value: 5.332e-08
```
---

### 7. Linear regression on plot

```r
ggplot(data = dat, aes(Education, Fertility, color = Catholic.bin, group = Catholic.bin)) + 
    geom_point() + geom_smooth(method = "lm")
```

![](slides_files/figure-html/data-vis-mod-1.png)

---

## Data journalism

[Source: SRF](https://www.srf.ch/static/srf-data/data/2018/federer/#/en)

---

## Data journalism

[Source: SRF](https://srfdata.github.io/2018-01-roger-federer/)

---

## Memes and GIFs

[Source: Danielle Navaro](http://djnavarro.net/post/2018-05-03-valid-social-commentary/)

---

## Memes and GIFs

```r
## install.packages('glue') ## run this once install.packages('meme') ## run
## this once

library(meme)  ## tell R that the meme package is needed
library(glue)  ## tell R that the glue package is needed

loc <- "https://djnavarro.net/img/meme/"
```

```r
meme(img = glue(loc, "morpheus.png"), upper = "what if i told you", lower = "i made this in R", 
    size = 1)
```

[Source: Danielle Navaro](http://djnavarro.net/post/2018-05-03-valid-social-commentary/)

---

## Memes and GIFs

```r
meme(img = "img/cat.jpeg", upper = "I have no time to be impressed", lower = "MMMkay", 
    size = 1)
```

![](slides_files/figure-html/meme2-1.png)

---

## Animations

<blockquote class="twitter-tweet" data-cards="hidden" data-lang="en">Getting ready to teach dplyr joins to new <a href="https://twitter.com/hashtag/rstats?src=hash&amp;ref_src=twsrc%5Etfw">#rstats</a> users tomorrow, so naturally I productively procrastinated by getting to know the new gganimate. It is the coolest! <a href="https://t.co/1kkOi5D5TK">pic.twitter.com/1kkOi5D5TK</a>&mdash; Garrick Aden-Buie (@grrrck) <a href="https://twitter.com/grrrck/status/1029567123029467136?ref_src=twsrc%5Etfw">August 15, 2018</a></blockquote>
<script async src="https://platform.twitter.com/widgets.js" charset="utf-8"></script>

---

## Animations

[Source: Github](https://github.com/gadenbuie/tidy-animated-verbs#tidy-animated-verbs)

---

## Decision making for lunch

<blockquote class="twitter-tweet" data-lang="en">Just learned, courtesy of <a href="https://twitter.com/AdamGruer?ref_src=twsrc%5Etfw">@AdamGruer</a>, that there is an <a href="https://twitter.com/hashtag/rstats?src=hash&amp;ref_src=twsrc%5Etfw">#rstats</a> compiler app on iOS. What people may not know is that for years I have been using &#39;sample&#39; function in R to make decisions for difficult choices. Now I have it handy in my mobile! <a href="https://t.co/vQxVNhjchm">https://t.co/vQxVNhjchm</a> <a href="https://t.co/rCPubC6BeD">pic.twitter.com/rCPubC6BeD</a>&mdash; Emi Tanaka 🌾 (@statsgen) <a href="https://twitter.com/statsgen/status/1027332304656465920?ref_src=twsrc%5Etfw">August 8, 2018</a></blockquote>
<script async src="https://platform.twitter.com/widgets.js" charset="utf-8"></script>

---
class: inverse, center, middle

# How can I use it?
---

class: left, middle
## Getting started

1. **Install** R :
  - On a computer: via [RStudio](https://www.rstudio.com/products/rstudio/download/) or [R project](https://stat.ethz.ch/CRAN/).

- Or (easier) use R in browser: [rdrr.io/snippets/](https://rdrr.io/snippets/) (no login required) or [Studio Cloud](https://rstudio.cloud/) (login with google or githu baccount).

1. ⁉️ Come up with a **question** you want to answer. 
--

1. Get your hands on **data** 🔀 Take part in [TidyTuesday](https://github.com/rfordatascience/tidytuesday).

---
class: inverse, center, middle

# R Community

---
class: center, middle

## R is developing quickly

## R community can help you learn!

---
class: left, middle

## [RWeekly](https://rweekly.org/) Newsletter

- Submit & subscribe here: https://rweekly.org/

- Weekly selection of **blogs** delivered into your mailbox.

- Replicating code is a good way to learn!

---
class: left, middle

## [TidyTuesday](https://github.com/rfordatascience/tidytuesday)

TidyTuesday provides you weekly with a new dataset (and a goal).

<blockquote class="twitter-tweet" data-lang="en">The <a href="https://twitter.com/R4DScommunity?ref_src=twsrc%5Etfw">@R4DScommunity</a> welcomes you to week 19 of <a href="https://twitter.com/hashtag/tidytuesday?src=hash&amp;ref_src=twsrc%5Etfw">#tidytuesday</a>! We&#39;re exploring <a href="https://twitter.com/FiveThirtyEight?ref_src=twsrc%5Etfw">@FiveThirtyEight</a> data on airline safety! Many thanks to 538 package maintainers! Data: <a href="https://t.co/sElb4fcv3u">https://t.co/sElb4fcv3u</a> Article: <a href="https://t.co/qmm69g8khc">https://t.co/qmm69g8khc</a> <a href="https://twitter.com/hashtag/r4ds?src=hash&amp;ref_src=twsrc%5Etfw">#r4ds</a> <a href="https://twitter.com/hashtag/tidyverse?src=hash&amp;ref_src=twsrc%5Etfw">#tidyverse</a> <a href="https://twitter.com/hashtag/rstats?src=hash&amp;ref_src=twsrc%5Etfw">#rstats</a> <a href="https://twitter.com/hashtag/dataviz?src=hash&amp;ref_src=twsrc%5Etfw">#dataviz</a> <a href="https://t.co/wl5UZCojEP">pic.twitter.com/wl5UZCojEP</a>&mdash; Thomas @ Strata Data NY - RStudio Booth (@thomas_mock) <a href="https://twitter.com/thomas_mock/status/1026505945722101760?ref_src=twsrc%5Etfw">August 6, 2018</a></blockquote>
<script async src="https://platform.twitter.com/widgets.js" charset="utf-8"></script>

---
class: left, middle

## TidyTuesday

.pull-left[
### Dataset

<a href="https://github.com/rfordatascience/tidytuesday/blob/master/data/2018-08-07/week19_airline_safety.csv">
<img src="img/airline_data.png" width="400">
</a>
]

.pull-right[

### Goal (if you need one)
<a href="https://fivethirtyeight.com/features/should-travelers-avoid-flying-airlines-that-have-had-crashes-in-the-past/">
<img src="img/airline_publication.png" width="300">
</a>

]

---
class: left, middle

## -Ladies

- **Global** organisation.

- **Mission**: *To increase gender diversity in the R community* by encouraging, inspiring, and empowering underrepresented minorities.

- Founded in 2012 by [**Gabriela de Queiroz**](https://rladies.org/united-states-rladies/name/gabriela-de-queiroz/).

- Currently **125 R-Ladies meetup groups** in 40 countries.

---
class: center, middle

.footnote[[Source code](https://github.com/rladies/Map-RLadies-Growing) by [Daniela Vázquez](https://twitter.com/d4tagirl).]

---
class: left, middle

- Find our more about **R-Ladies**: https://rladies.org/

- Find other **chapters**: https://gqueiroz.shinyapps.io/rshinylady/

- Find **speakers**: https://rladies.org/directory/

---
class: left, middle

## R user groups nearby

- R-Ladies chapter in Lausanne: https://www.meetup.com/rladies-lausanne/

- Geneve R user group: https://www.meetup.com/Geneve-R-User-Group/

- R Lunches in Geneve: http://use-r-carlvogt.github.io/prochains-lunchs/

- adminR in Bern: https://www.meetup.com/adminR/

- Check out global list [here](https://jumpingrivers.github.io/meetingsR/) (provided by jumping rivers)

---
class: left, middle

## Keeping up on the road: Podcasts

- [Not So Standard Deviations](https://soundcloud.com/nssd-podcast) by Hilary Parker and Roger Peng. Data science podcast.

- [Credibly Curious](https://soundcloud.com/crediblycurious) by Saskia Freytag and Nicholas Tierney. A podcast about R and statistics.

- [DataCamp Podcast](https://www.datacamp.com/community/podcast) explores different data science jobs. It's not always, but often about R.

---
class: inverse, center, middle

# Thank you!

Slides: [https://sinarueeger.github.io/20180912-geek-girls-carrots/slides#1](https://sinarueeger.github.io/20180912-geek-girls-carrots/slides#1)

Source code: [https://github.com/sinarueeger/20180912-geek-girls-carrots/](https://github.com/sinarueeger/20180912-geek-girls-carrots/)

: [@sinarueeger](https://twitter.com/sinarueeger)