Laboratorio Bio-demografico

class: center, middle, inverse, title-slide

# Laboratorio Bio-demografico
## Lezione 3
### Nicola Barban <br> Alma Mater Studiorum Università di Bologna <br> Dipartimento di Scienze Statistiche
### 18 Febbraio 2021<br> <img src="../img/UniBo-Universita-di-Bologna.png" style="width:20.0%" />

---

# Outline
1. Visualizzare quantità con ggplot2
2. Visualizzare relazioni tra variabili 
3. Opendata COVID-19. Primi passi

### Pacchetti da scaricare

* `library(ggplot2)`
* `library(gapminder)`
* `library(forcats)`

### Cheatsheet ggplot2
https://rstudio.com/wp-content/uploads/2016/11/ggplot2-cheatsheet-2.1.pdf
---

# US income data

```r
# importing data
US_income<-read.csv("US_income.csv")

# listing objects
ls()
```

```
## [1] "US_income"
```

```r
# prime 6 righe di US_income
head(US_income)
```

```
##   X GEOID        name median_income median_income_moe population         area
## 1 1     1     Alabama         43623               281    4830620 133958437749
## 2 2     4     Arizona         50255               211    6641928 295232708152
## 3 3     5    Arkansas         41371               247    2958208 137792577218
## 4 4     6  California         61818               156   38421464 410516610493
## 5 5     8    Colorado         60629               252    5278906 269580118211
## 6 6     9 Connecticut         70331               409    3593222  12961831628
##        popdens
## 1 3.606059e-05
## 2 2.249726e-05
## 3 2.146856e-05
## 4 9.359296e-05
## 5 1.958196e-05
## 6 2.772156e-04
```

---

# il mio primo ggplot

```r
# loading ggplot2 library
library(ggplot2)

# Basic barplot
plot1 <- ggplot(data=US_income, aes(x=name, y=median_income)) +
*       geom_bar(stat="identity")
```
---

```r
plot1
```

<img src="lezione3_files/figure-html/unnamed-chunk-1-1.png" width="50%" />
---
# Cambia coordinate

```r
#Horizontal bar plot

plot1 + coord_flip()
```

---
# Anatomia di un ggplot

1. load library (solo una volta) `library(ggplot2)`
2. define data `ggplot(data=US_income,`
3. define aesthetics `aes(x=name, y=median_income)` *links variables to things you will see *
4. add a layer to the graph `+geom_bar(`
5. define layer specific aesthetics `stat="identity"` *identity non calcola nessuna funzione*
6. Save into an object `plot1<-` e plot `plot1`
7. add another layer `+ coord_flip()`

---
# Esercizio 1
* Riproduci lo stesso grafico per la popolazione

---
# Soluzione 1

```r
library(ggplot2)
# Basic barplot
plot2<-ggplot(data=US_income, aes(x=name, y=population)) +
    geom_bar(stat="identity")+ coord_flip()
```

---

```r
plot2
```

---
# Dotplot

```r
# Basic barplot
dotplot<-ggplot(data=US_income, aes(x=name, y=population)) +
     geom_point()+ coord_flip()
```
---

```r
dotplot
```

---
# Cambia colore

```r
# Basic barplot
p3<-ggplot(data=US_income, aes(x=name, y=median_income)) +
*   geom_bar(stat="identity", color="blue", fill="blue") +
    coord_flip()
```
---

```r
p3
```

---
# Ggplot theme

```r
# Basic barplot
p4<-ggplot(data=US_income, aes(y=name, x=median_income)) +
    geom_bar(stat="identity", fill="blue", color="blue") +
*   theme_minimal()
p4
```

<img src="lezione3_files/figure-html/income-by-state-color-minimal-1.png" width="50%" />
---
# Ggplot themes
https://ggplot2-book.org/polishing.html

* `theme_bw()`: a variation on theme_grey() that uses a white background and thin grey grid lines.

* `theme_linedraw()`: A theme with only black lines of various widths on white backgrounds, reminiscent of a line drawing.

* `theme_light()`: similar to theme_linedraw( ) but with light grey lines and axes, to direct more attention towards the data.

* `theme_dark()`: the dark cousin of theme_light(), with similar line sizes but a dark background. Useful to make thin coloured lines pop out.

* `theme_minimal()`: A minimalistic theme with no background annotations.

* `theme_classic()`: A classic-looking theme, with x and y axis lines and no gridlines.

* `theme_void()` : A completely empty theme.
---
# Altri themes

```r
install.packages('ggthemes', dependencies = TRUE)
```

```
## 
## The downloaded binary packages are in
## 	/var/folders/9l/47_t81dd0g32mw3bc16019dr0000gn/T//Rtmp0FP8NL/downloaded_packages
```

```r
library(ggthemes)
plot2+  theme_economist()
```

<img src="lezione3_files/figure-html/unnamed-chunk-6-1.png" width="50%" />
---

---
# Riordina i fattori

```r
*library(forcats)
help(fct_reorder)
# Basic barplot
p5<-ggplot(data=US_income, aes(y=fct_reorder(name,median_income), x=median_income)) +
    geom_bar(stat="identity", fill="blue", color="blue") +
    theme_minimal()
p5
```

<img src="lezione3_files/figure-html/unnamed-chunk-7-1.png" width="50%" />
---

# Aggiungi etichette sugli assi

```r
p6<-p5+
* xlab("Reddito mediano")+
  ylab("Stato")
p6
```

---
# Esercizio 2
* Fai  un dotplot  per popolazione, punti colorati di rosso
---
# Soluzione

```r
p7<-ggplot(data=US_income, aes(y=fct_reorder(name,population), x=population)) +
    geom_point( color="red") +
    xlab("Population")+
    ylab("Country")+
  theme_minimal()
  p7
```

---
# Aggiungi titolo

```r
*p7+ggtitle("Popolazione Stati Uniti")
```

---

# Gapminder

```r
library(gapminder)
data(gapminder)
help(gapminder)
```
---

# Scatterplot

```r
g1<-ggplot( data=gapminder, aes(y=lifeExp, x= gdpPercap))+
    geom_point(color="red")+
    theme_minimal()+
    ylab("Aspettativa di vita")+
    xlab("PIL pro capite")
g1
```

<img src="lezione3_files/figure-html/unnamed-chunk-12-1.png" width="50%" />
---

# Scatterplot

```r
g2<-g1+geom_smooth(method="lm")
g2
```

<img src="lezione3_files/figure-html/unnamed-chunk-13-1.png" width="50%" />
---

# Scatterplot con scala logaritmica (in base 10)

```r
g3<-g1+
    scale_x_log10()+
*   geom_smooth(method="lm")
g3
```

<img src="lezione3_files/figure-html/unnamed-chunk-14-1.png" width="50%" />
---

# Scatterplot con scala logaritmica

```r
g3<-g1+
    scale_x_log10(labels=scales::dollar)+
    geom_smooth(method="lm")+
  ggtitle("Aspettativa di vita e PIL")
g3
```

---
# Esercizio 3
* riprodurre il grafico precedente con le seguenti caratteristiche:
  + selezionare solo dati di paesi provenienti dal continente "Asia"
  + attribuire alla linea di regressione il colore `orange`
  + rimuovere dalla linea di regressione l'intervallo di confidenza
---
# Soluzione 3

```r
es2<-ggplot( data=subset(gapminder, continent=="Asia"), aes(y=lifeExp, x= gdpPercap))+
    geom_point(color="red")+
    theme_minimal()+
    ylab("Aspettativa di vita")+
    xlab("PIL pro capite")+
    scale_x_log10(labels=scales::dollar)+
    geom_smooth(method="lm", color="orange", se=F)+
  ggtitle("Aspettativa di vita e PIL")
```

---

```r
es2
```

---
# Versione più accurata

```r
gapm<-ggplot( data=gapminder, aes(y=lifeExp, x= gdpPercap))+
    geom_point(alpha=0.3)+
*   scale_x_log10(labels=scales::dollar)+
    geom_smooth(method="lm", color="orange", se=F)+
    labs( y="Aspettativa di vita",
          x="PIL pro capite",
          title="Aspettativa di vita e PIL",
          subtitle = "Ogni punto rappresenta un paese in un anno",
          caption= "Fonte: Gapminder")   
```

---

```r
gapm
```

<img src="lezione3_files/figure-html/unnamed-chunk-19-1.png" width="50%" />
---
# Fit nonparametrico

```r
gapm.np<-ggplot( data=gapminder, aes(y=lifeExp, x= gdpPercap))+
    geom_point(alpha=0.3)+
    scale_x_log10(labels=scales::dollar)+
    geom_smooth(method="loess", color="orange", se=F)+
    labs( y="Aspettativa di vita",
          x="PIL pro capite",
          title="Aspettativa di vita e PIL",
          subtitle = "Ogni punto rappresenta un paese in un anno",
          caption= "Fonte: Gapminder")
```

---

```r
gapm.np
```

---
# Grafico per categorie

```r
ggplot( data=gapminder, aes(y=lifeExp, x= gdpPercap, col=continent))+
    geom_point(alpha=0.3)+
    scale_x_log10(labels=scales::dollar)+
    geom_smooth(method="loess")+
  theme_minimal()
```

<img src="lezione3_files/figure-html/unnamed-chunk-22-1.png" width="50%" />
---

# Covid-19 Opendata Vaccini
https://github.com/italia/covid19-opendata-vaccini

---
# Dati vaccinazioni COVID Italia

```r
vaccini_anagrafica <-read.csv("https://raw.githubusercontent.com/italia/covid19-opendata-vaccini/master/dati/anagrafica-vaccini-summary-latest.csv",  header=T)

vaccini1<-ggplot(data=vaccini_anagrafica, aes(x=fascia_anagrafica, y=totale/1000)) +
    geom_bar(stat="identity", fill="red") +
    xlab("Fascia Età")+
    ylab("Dosi somministrate (migliaia)")+
  theme_minimal()
```
---

```r
vaccini1
```

<img src="lezione3_files/figure-html/unnamed-chunk-23-1.png" width="50%" />
---

```r
vaccini_somministrazione <-read.csv("https://raw.githubusercontent.com/italia/covid19-opendata-vaccini/master/dati/vaccini-summary-latest.csv",  header=T)

v2<-ggplot(data=subset(vaccini_somministrazione,area!="ITA"),
       aes(y=fct_reorder(area,dosi_somministrate,sum), x=dosi_somministrate/1000)) +
    geom_bar(stat="identity", fill="darkgreen") +
    ylab("Regione")+
    xlab("Dosi somministrate (migliaia)")+
  theme_minimal()
```

---

```r
v2
```

---
# Scarica dati somministrazione nel tempo

```r
vaccini_somministrazione_summary <-read.csv("https://raw.githubusercontent.com/italia/covid19-opendata-vaccini/master/dati/somministrazioni-vaccini-summary-latest.csv",  header=T)
dim(vaccini_somministrazione_summary)
```

```
## [1] 1045   15
```
---

# Formato data

```r
str(vaccini_somministrazione_summary)
```

```
## 'data.frame':	1045 obs. of  15 variables:
##  $ data_somministrazione                     : chr  "2021-01-08" "2021-02-01" "2021-02-01" "2021-01-17" ...
##  $ area                                      : chr  "LIG" "PIE" "UMB" "ABR" ...
##  $ totale                                    : int  2685 8637 1260 187 349 713 9443 469 305 301 ...
##  $ sesso_maschile                            : int  927 3051 333 59 93 300 3073 191 119 110 ...
##  $ sesso_femminile                           : int  1758 5586 927 128 256 413 6370 278 186 191 ...
##  $ categoria_operatori_sanitari_sociosanitari: int  1911 6161 876 136 318 656 5757 397 277 301 ...
##  $ categoria_personale_non_sanitario         : int  429 1402 32 51 15 57 2043 72 27 0 ...
##  $ categoria_ospiti_rsa                      : int  345 1074 352 0 16 0 1643 0 1 0 ...
##  $ categoria_over80                          : int  0 0 0 0 0 0 0 0 0 0 ...
##  $ prima_dose                                : int  2685 577 47 180 227 713 654 469 1 301 ...
##  $ seconda_dose                              : int  0 8060 1213 7 122 0 8789 0 304 0 ...
##  $ codice_NUTS1                              : chr  "ITC" "ITC" "ITI" "ITF" ...
##  $ codice_NUTS2                              : chr  "ITC3" "ITC1" "ITI2" "ITF1" ...
##  $ codice_regione_ISTAT                      : int  7 1 10 13 4 3 1 4 6 10 ...
##  $ nome_area                                 : chr  "Liguria" "Piemonte" "Umbria" "Abruzzo" ...
```

```r
vaccini_somministrazione_summary$giorno<-as.Date(vaccini_somministrazione_summary$data_somministrazione)
```
---

# Line plot

```r
v3<-ggplot(data=subset(vaccini_somministrazione_summary, area=="EMR"),
           aes(x=giorno, y=totale))+ylab("Dosi somministrate ")
```

---

```r
v3+geom_line(color="#69b3a2")+theme_minimal()
```

<img src="lezione3_files/figure-html/unnamed-chunk-27-1.png" width="50%" />
---

# Esercizio 4
* Mostra l'andamento della campagna vaccinale nelle  regioni:
  + Emilia Romagna
  + Lombardia
  + Veneto
  + Campania
---

# Soluzione 4

```r
sol4<-ggplot(data=subset(vaccini_somministrazione_summary,
                  area==("EMR") |
                  area==("LOM") |
                  area==("VEN") |
                  area==("CAM") ),
           aes(x=giorno, y=totale))+
           geom_line(aes(group=area, color=area))+
  theme_minimal()
```
---

```r
sol4
```

<img src="lezione3_files/figure-html/unnamed-chunk-29-1.png" width="50%" />
---
# Soluzione 3 alternativa

```r
regioni<-ggplot(data=subset(vaccini_somministrazione_summary,
                  area==("EMR") |
                  area==("LOM") |
                  area==("VEN") |
                  area==("CAM") ),
           aes(x=giorno, y=totale))+
*          geom_line()+facet_wrap(~area)
*#alt+125 ~
```

---

```r
regioni+ylab("Dosi somministrate ")
```

---
# Salva i grafici in un documento PDF

```r
ggsave("ilmiografico.pdf", plot=regioni)
```