Tutorial: Use Cases
tutorial.Rmd1. Installing LHTpicker package
For this vigenette to work LHTpicker must be installed
and exist in R’s default libpath.
# devtools::install_github("https://github.com/d2gex/LHTPicker", dep=TRUE)2. Fetch wanted LHTs from FishLife
There are two use cases for using FishLife to obtain missing Life History Traits (LHTs): i) You want to evaluate a species for which no biological parameters are available, meaning you need to retrieve FishLife-predicted LHTs for the desired taxon. ii) You have some biological parameters but not all, and you use FishLife to generate a new set of parameters that integrates and updates your existing data.
This section deals with case i.
2.1 Inputted taxa details
A csv file with the taxon names and the wanted LHTs you are seeking is required.The csv shows as columns those LHTs you are after, as shown below.
taxon_lhts_to_fetch <- readRDS("data/wanted_taxon_details.rds")
head(taxon_lhts_to_fetch)
#> taxon Linf Winf K L50 M Amat Amax Temperature
#> 1 Trisopterus luscus NA NA NA NA NA NA NA NA
#> 2 Pollachius pollachius NA NA NA NA NA NA NA NAThe wrapper uses a list to map column names in your csv to the actual
names used by FishLife. Should some predicted variable names change in
FishLife in the future, you may need to tweak the list. See that such a
list is a member of an R6 object and it is called
fishlife_context. If nothing has changed in FishLife,
unless you need to add more wanted parameters, nothing else is required
to do.
LHTpicker::fishlife_context$lht_names
#> $Linf
#> [1] "log(length_infinity)"
#>
#> $Winf
#> [1] "log(weight_infinity)"
#>
#> $K
#> [1] "log(growth_coefficient)"
#>
#> $M
#> [1] "log(natural_mortality)"
#>
#> $L50
#> [1] "log(length_maturity)"
#>
#> $Amax
#> [1] "log(age_max)"
#>
#> $Amat
#> [1] "log(age_maturity)"
#>
#> $Temperature
#> [1] "temperature"Likewise, to back-transform the obtained predicted variables
LHTpicker provides a function that map such predicted
variables with the required function to convert it. As previously
mentioned, you may not need to touch this list either.
LHTpicker::fishlife_context$backtransform_function_list
#> $`log(length_infinity)`
#> function (x) .Primitive("exp")
#>
#> $`log(weight_infinity)`
#> function (x) .Primitive("exp")
#>
#> $`log(growth_coefficient)`
#> function (x) .Primitive("exp")
#>
#> $`log(natural_mortality)`
#> function (x) .Primitive("exp")
#>
#> $`log(length_maturity)`
#> function (x) .Primitive("exp")
#>
#> $`log(age_max)`
#> function (x) .Primitive("exp")
#>
#> $`log(age_maturity)`
#> function (x) .Primitive("exp")
#>
#> $temperature
#> function (x)
#> x
#> <bytecode: 0x555c39c6b860>
#> <environment: namespace:base>2.2 Fetching predicted LHTs
Please refer to the reference for further details.
p_lht_picker <- LHTpicker::PredictedLHTPicker$new(FishLife::FishBase_and_Morphometrics,
LHTpicker::fishlife_context$lht_names,
LHTpicker::fishlife_context$backtransform_function_list,
taxon_lhts_to_fetch)
predicted_lht_df <- p_lht_picker$pick_and_backtransform()
head(predicted_lht_df)
#> taxon Linf Winf K L50 M
#> 1 Trisopterus luscus 43.86770 833.9878 0.3737931 19.52203 0.5981946
#> 2 Pollachius pollachius 87.30785 5819.9735 0.1867131 34.71789 0.3085845
#> Amat Amax Temperature
#> 1 1.400562 6.458926 17.36961
#> 2 3.456610 12.138258 12.11773In case FishLife cannot provide details about a species for whatever
reason, LHTpicker will return a column of NA values for the
rows holding details about the missing species, as shown below.
non_existant_taxon_lhts_to_fetch <- dplyr::mutate(taxon_lhts_to_fetch, taxon = dplyr::case_when(
taxon == "Trisopterus luscus" ~ "IDoNoExist",
.default = taxon
))
p_lht_picker <- LHTpicker::PredictedLHTPicker$new(FishLife::FishBase_and_Morphometrics,
LHTpicker::fishlife_context$lht_names,
LHTpicker::fishlife_context$backtransform_function_list,
non_existant_taxon_lhts_to_fetch)
predicted_lht_df <- p_lht_picker$pick_and_backtransform()
head(predicted_lht_df)
#> # A tibble: 2 × 9
#> taxon Linf Winf K L50 M Amat Amax Temperature
#> <chr> <dbl> <dbl> <dbl> <dbl> <dbl> <dbl> <dbl> <dbl>
#> 1 IDoNoExist NA NA NA NA NA NA NA NA
#> 2 Pollachius pollachius 87.3 5820. 0.187 34.7 0.309 3.46 12.1 12.13 Fetch updated LHTs given new data
This section deals with case ii, that is, with fetching updated LHTs from FishLife.
3.1 Inputted predicting taxa details
A csv file with the taxon names you are after and the wanted-to-update LHTs is required as shown below. Let’s suppose we have most of the LHTs however we do not know the natural mortality(M) nor the age of maturity.
taxon_lhts_to_update <- readRDS("data/wanted_update_taxon_details.rds")
head(taxon_lhts_to_update)
#> # A tibble: 2 × 9
#> taxon Linf Winf K L50 M Amat Amax Temperature
#> <chr> <dbl> <dbl> <dbl> <dbl> <lgl> <lgl> <dbl> <dbl>
#> 1 Trisopterus luscus 42.4 921 0.21 19.4 NA NA 9 14.3
#> 2 Pollachius pollachius 102. 12045 0.193 41.6 NA NA 8 14.33.2 Fetching updated LHTs
Unless the previous function, we need to provide two extra
parameters: a prefix that LHTpicker will use to concatenate
to the named of updated columns in the csv and function transformation
list, similar to the back-transformation one, to transform the provided
LHTs into Fishlife’s expected domain. Once again, unless you need other
variables from FishLife, you do not need to modify neither of the
related transformation and back-transformation lists.
u_lht_picker <- LHTpicker::UpdatedLHTPicker$new(
FishLife::FishBase_and_Morphometrics,
taxon_lhts_to_update,
LHTpicker::fishlife_context$updated_prefix,
LHTpicker::fishlife_context$transform_function_list,
LHTpicker::fishlife_context$backtransform_function_list,
LHTpicker::fishlife_context$lht_names
)
updated_lht_df <- u_lht_picker$pick_and_backtransform()
head(updated_lht_df)
#> taxon Linf Winf K L50 M Amat Amax Temperature
#> 1 Trisopterus luscus 42.41 921 0.210 19.45 NA NA 9 14.3
#> 2 Pollachius pollachius 102.14 12045 0.193 41.60 NA NA 8 14.3
#> updated_Linf updated_Winf updated_K updated_M updated_L50 updated_Amax
#> 1 44.01880 839.5509 0.3451741 0.5872462 19.60257 6.973144
#> 2 95.04985 8387.9049 0.1881105 0.2997824 36.48049 11.111069
#> updated_Amat updated_Temperature
#> 1 1.438878 16.75299
#> 2 3.413651 13.05912See how each updated column name from FishLife has a prefix attached
to it and how updated_M and Amat have now been
provided. In order to use the these LHTs for your assessment you need to
use the whole returned set, to avoid breaking the covariance matrix
among parameters (Thorson et al., 2017)