Data science with {hyenaR}:
LESSON 9

Prepare our workspace


STEP 1: Load required packages

library(hyenaR) ## For our hyena specific functions
library(dplyr) ## For most data wrangling
library(ggplot2) ## For plotting
library(lubridate) ## Working with dates
library(tidyr) ## Extra data wrangling functions
library(stringr) ## Working text


STEP 2: Load the database

load_package_database.full(
  
  # Location of our database file
  db.path = "example_git/source_data/Fisidata_2022_08_10.sqlite"
  
)

Today’s goals

GOAL 1: create_sample_starting.table()


GOAL 2: Working with text columns

create_sample_starting.table

Return sample records


All samples for a given ID

create_sample_starting.table(ID = "A-010")
# A tibble: 2 × 5
  ID    sampleID collection_time     type   treatment
  <chr> <chr>    <dttm>              <chr>  <chr>    
1 A-010 N0386a   1999-12-01 14:30:00 biopsy liq N    
2 A-010 N0386b   1999-12-01 14:30:00 biopsy Ethanol

Return sample records


All samples for a date range

create_sample_starting.table(from = "1999-01-01", to = "2000-01-01")
# A tibble: 177 × 5
   ID    sampleID collection_time     type     treatment
   <chr> <chr>    <dttm>              <chr>    <chr>    
 1 M-052 N0340a   1999-04-04 12:55:00 faeces   liq N    
 2 M-052 N0340b   1999-04-04 12:55:00 faeces   liq N    
 3 M-052 N0340c   1999-04-04 12:55:00 mucus    liq N    
 4 M-052 N0340d   1999-04-04 12:55:00 mucus    liq N    
 5 M-052 N0340e   1999-04-04 12:55:00 mucus    Ethanol  
 6 M-123 N0341a   1999-04-11 11:15:00 faeces   liq N    
 7 M-123 N0341b   1999-04-11 11:15:00 faeces   liq N    
 8 M-123 N0341c   1999-04-11 11:15:00 mucus    liq N    
 9 M-123 N0341d   1999-04-11 11:15:00 parasite 10% Form 
10 M-123 N0341e   1999-04-11 11:15:00 mucus    liq N    
# … with 167 more rows

Using stringr to filter text columns


Simple, Consistent Wrappers for Common String Operations


‘string’ = ‘character string’

Using stringr to filter text columns


Identify samples from liver

create_sample_starting.table() %>% 
  mutate(is_liver = stringr::str_detect(string = type,
                                        pattern = "liver"))
# A tibble: 9,029 × 6
   ID    sampleID collection_time     type           treatment is_liver
   <chr> <chr>    <dttm>              <chr>          <chr>     <lgl>   
 1 M-013 N0000a   1996-07-22 18:50:00 faeces         liq N     FALSE   
 2 M-013 N0000b   1996-07-22 18:50:00 faeces         liq N     FALSE   
 3 L-043 N0001    1996-07-30 16:30:00 testicle       liq N     FALSE   
 4 L-043 N0002a   1996-07-30 14:24:00 liver          10% Form  TRUE    
 5 L-043 N0002    1996-07-30 16:30:00 liver          liq N     TRUE    
 6 L-043 N0003a   1996-07-30 14:30:00 spleen         10% Form  FALSE   
 7 L-043 N0003    1996-07-30 16:30:00 spleen         liq N     FALSE   
 8 L-043 N0004a   1996-07-30 15:30:00 salivary gland 10% Form  FALSE   
 9 L-043 N0004    1996-07-30 16:30:00 salivary gland liq N     FALSE   
10 L-043 N0005a   1996-07-30 15:30:00 lymph node     10% Form  FALSE   
# … with 9,019 more rows

Using stringr to filter text columns


Count liver samples per year

create_sample_starting.table() %>% 
  filter(stringr::str_detect(string = type,
                             pattern = "liver")) %>% 
  mutate(year = lubridate::year(collection_time)) %>% 
  group_by(year) %>% 
  count() %>% 
  ungroup()
# A tibble: 16 × 2
    year     n
   <dbl> <int>
 1  1996     2
 2  1997     6
 3  1998     4
 4  2002     5
 5  2003     2
 6  2004     1
 7  2005     3
 8  2006     3
 9  2007     4
10  2008     2
11  2011     1
12  2012     4
13  2017     2
14  2018     4
15  2019     5
16    NA    18

Using stringr to filter text columns


Count liver OR kidney samples per year

create_sample_starting.table() %>% 
  filter(stringr::str_detect(string = type,
                             pattern = "liver|kidney")) %>% 
  mutate(year = lubridate::year(collection_time)) %>% 
  group_by(year) %>% 
  count() %>% 
  ungroup()
# A tibble: 16 × 2
    year     n
   <dbl> <int>
 1  1996     2
 2  1997     8
 3  1998     6
 4  2002     9
 5  2003     4
 6  2004     1
 7  2005     5
 8  2006     5
 9  2007     7
10  2008     6
11  2011     3
12  2012     8
13  2017     3
14  2018     8
15  2019     8
16    NA    28

Using stringr to filter text columns


Count number of unique individuals from which we have brain samples

create_sample_starting.table()

Using stringr to filter text columns


Count number of unique individuals from which we have brain samples

create_sample_starting.table() %>% 
  filter(stringr::str_detect(string = type,
                             pattern = "brain")) %>% 
  group_by(ID) %>% 
  summarise() %>% 
  nrow()

Using stringr to filter text columns


Find records of strep or disease in injuries table

extract_database_table("injuries") %>% 
  filter(stringr::str_detect(tolower(cause),
                             pattern = "strep|disease")) %>% 
  mutate(year = lubridate::year(date_time)) %>% 
  group_by(year) %>% 
  summarise(n_disease = n())
# A tibble: 23 × 2
    year n_disease
   <dbl>     <int>
 1  1998         3
 2  1999         2
 3  2000         1
 4  2001         2
 5  2002        15
 6  2003         6
 7  2004         5
 8  2005         5
 9  2006         8
10  2007         4
# … with 13 more rows

Using stringr to filter text columns


Find records of strep or disease in injuries table

Code 
plot_data <- extract_database_table("injuries") %>% 
  filter(stringr::str_detect(tolower(cause), pattern = "strep|disease")) %>% 
  mutate(year = lubridate::year(date_time)) %>% 
  group_by(year) %>% 
  summarise(n_disease = n())

ggplot(data = plot_data) +
  geom_col(aes(x = year, y = n_disease)) +
  labs(y = "Number of diseased individuals sighted",
       title = "Number of disease sightings per year",
       subtitle = "Includes sightings caused by both streptococcus and other disease") +
  scale_x_continuous(limits = c(1996, NA)) +
  theme_classic() +
  theme(axis.title.x = element_blank())

Using stringr to filter text columns


Other useful functions

## Replace character with something new
stringr::str_replace_all(string = "liver_kidney/brain",
                         pattern = "_|/",
                         replacement = ", ")
[1] "liver, kidney, brain"
## Remove unwanted characters
stringr::str_remove_all(string = "sample_1_A",
                        pattern = "_")
[1] "sample1A"
## Remove unwanted white space
stringr::str_trim(string = "     new den  ")
[1] "new den"

HOMEWORK: The same as last week!

TASK 1:

Use create_id_starting.table() to find all individuals born in main clans in the years 1997 - 2021 (where we have full year observation).


How many individuals have been born in total during this period?

TASK 2:

Extract individual birth clan, birth date, sex, and lifespan.


How many individuals have missing data?


How would you interpet NAs in each column?

TASK 3:

Extract the year of birth for each individual.


In which year were the most cubs born?


Is the most productive year the same for all clans?


BONUS: Use filter to return a data frame with only the best year(s) for each clan?

TASK 4:

Extract the month of birth for each individual.


What was the most productive month in the crater from 1997-2021?


How many cubs were produced in that month?