Error flagged with as_epidist for epireview object

[adamkucharski]

Please place an "x" in all the boxes that apply

• I have the most recent version of this package and R
• I have found a bug
• I have a reproducible example
• I want to request a new feature

---

The below call to as_epidist with a Lassa gamma distributed parameter is returning the following error:
Error in is_epidist_params(prob_dist, prob_dist_params) : Assertion on 'prob_dist_params' failed: Must have unique names, but element 3 is duplicated.

lassa_data <- epireview::load_epidata("lassa")
lassa_params <- lassa_data$params
pick_gamma <- lassa_params[261,]
param1 <- as_epidist(pick_gamma)

---

[adamkucharski]

Also found an error with Ebola parameter with exponential distribution:

# Load Ebola data
data_in <- load_epidata(pathogen = "ebola")

# Extract parameters for onset to death, with qualtiy score > 50
params_in <- data_in[["params"]] 
params_infectious <- params_in |> dplyr::filter(parameter_type=="Human delay - infectious period") 

# Select distribution
params_inf_with_dist <- params_infectious |> dplyr::filter(!is.na(distribution_type))

# Start by extracting only those with Exponential
as_epidist(params_inf_with_dist)

[jfunction]

Looks like when

epiparameter/R/coercion.R

Line 128 in 7f69ecf

epidist <- epireview_to_epidist(x, ...)

gets called, eventually new_epidist gets called and

epiparameter/R/epidist.R

Line 68 in 7f69ecf

mean = summary_stats$mean

makes a new vector with 3 elements. Ie, a vector is created where mean is represented twice.

[joshwlambert]

Thanks for logging this issue. I've implemented a fix to these issues in #334.

Here are reproducible examples of the same code as above using the updated {epiparameter} functions.

Lassa example

library(epireview)
#> Loading required package: epitrix
#> Loading required package: ggplot2
#> Loading required package: ggforce

library(epiparameter)
lassa_data <- epireview::load_epidata("lassa")
#> Warning: One or more parsing issues, call `problems()` on your data frame for details,
#> e.g.:
#>   dat <- vroom(...)
#>   problems(dat)
#> One or more parsing issues, call `problems()` on your data frame for details,
#> e.g.:
#>   dat <- vroom(...)
#>   problems(dat)
#> One or more parsing issues, call `problems()` on your data frame for details,
#> e.g.:
#>   dat <- vroom(...)
#>   problems(dat)
#> Data loaded for lassa

lassa_params <- lassa_data$params
pick_gamma <- lassa_params[261,]
param1 <- as_epidist(pick_gamma)
#> Using Akhmetzhanov (2019). "<title not available>." _<journal not
#> available>_. 
#> To retrieve the citation use the 'get_citation' function
#> Warning: Cannot create full citation for epidemiological parameters without bibliographic information 
#>  see ?as_epidist for help.

param1
#> Disease: Lassa fever
#> Pathogen: Lassa mammarenavirus
#> Epi Distribution: human delay   incubation period
#> Study: Akhmetzhanov (2019). "<title not available>." _<journal not
#> available>_.
#> Distribution: gamma
#> Parameters:
#>   shape: 7.743
#>   scale: 1.653

unclass(param1)
#> $disease
#> [1] "Lassa fever"
#> 
#> $pathogen
#> [1] "Lassa mammarenavirus"
#> 
#> $epi_dist
#> [1] "Human delay - incubation period"
#> 
#> $prob_dist
#> <distribution[1]>
#> [1] Γ(7.7, 0.6)
#> 
#> $uncertainty
#> $uncertainty$shape
#> $uncertainty$shape$ci_limits
#> [1] NA
#> 
#> $uncertainty$shape$ci
#> [1] NA NA
#> 
#> $uncertainty$shape$ci_type
#> [1] NA
#> 
#> 
#> $uncertainty$scale
#> $uncertainty$scale$ci_limits
#> [1] NA
#> 
#> $uncertainty$scale$ci
#> [1] NA NA
#> 
#> $uncertainty$scale$ci_type
#> [1] NA
#> 
#> 
#> 
#> $summary_stats
#> $summary_stats$mean
#> [1] 12.8
#> 
#> $summary_stats$mean_ci_limits
#> [1] NA NA
#> 
#> $summary_stats$mean_ci
#> [1] NA
#> 
#> $summary_stats$sd
#> [1] 4.6
#> 
#> $summary_stats$sd_ci_limits
#> [1] NA NA
#> 
#> $summary_stats$sd_ci
#> [1] NA
#> 
#> $summary_stats$median
#> [1] NA
#> 
#> $summary_stats$median_ci_limits
#> [1] NA NA
#> 
#> $summary_stats$median_ci
#> [1] NA
#> 
#> $summary_stats$quantiles
#> [1] NA
#> 
#> $summary_stats$range
#> [1] NA NA
#> 
#> 
#> $citation
#> Akhmetzhanov (2019). "<title not available>." _<journal not
#> available>_.
#> 
#> $metadata
#> $metadata$sample_size
#> [1] NA
#> 
#> $metadata$region
#> [1] "Nigeria"
#> 
#> $metadata$transmission_mode
#> [1] NA
#> 
#> $metadata$vector
#> [1] NA
#> 
#> $metadata$extrinsic
#> [1] FALSE
#> 
#> $metadata$inference_method
#> [1] NA
#> 
#> 
#> $method_assess
#> $method_assess$censored
#> [1] NA
#> 
#> $method_assess$right_truncated
#> [1] NA
#> 
#> $method_assess$phase_bias_adjusted
#> [1] NA
#> 
#> 
#> $notes
#> [1] "No additional notes"
#> 
#> attr(,".epiparameter_namespace")
#> function () 
#> NULL
#> <bytecode: 0x114b79658>
#> <environment: namespace:epiparameter>

as.data.frame(lassa_params[261, ])
#>                                 id                parameter_data_id
#> 1 c6cead4f4e9802343ccbb00449f22471 e93361e40a2f1a00c337a420be54deb2
#>   covidence_id             pathogen                  parameter_type
#> 1         2617 Lassa mammarenavirus Human delay - incubation period
#>   parameter_value exponent parameter_unit parameter_lower_bound
#> 1            12.8        0           Days                    NA
#>   parameter_upper_bound parameter_value_type parameter_uncertainty_single_value
#> 1                    NA                 Mean                                 NA
#>   parameter_uncertainty_singe_type parameter_uncertainty_lower_value
#> 1                             <NA>                                NA
#>   parameter_uncertainty_upper_value parameter_uncertainty_type
#> 1                                NA                       <NA>
#>   cfr_ifr_numerator cfr_ifr_denominator distribution_type
#> 1                NA                  NA             Gamma
#>   distribution_par1_value distribution_par1_type distribution_par1_uncertainty
#> 1                    12.8                   Mean                          TRUE
#>   distribution_par2_value distribution_par2_type distribution_par2_uncertainty
#> 1                     4.6     Standard deviation                          TRUE
#>   method_from_supplement method_moment_value cfr_ifr_method method_r
#> 1                  FALSE             Endemic           <NA>     <NA>
#>   method_disaggregated_by method_disaggregated method_disaggregated_only
#> 1                    <NA>                FALSE                     FALSE
#>   riskfactor_outcome riskfactor_name riskfactor_occupation
#> 1               <NA>            <NA>                  <NA>
#>   riskfactor_significant riskfactor_adjusted population_sex
#> 1                   <NA>                <NA>    Unspecified
#>   population_sample_type            population_group population_age_min
#> 1         Hospital based Persons under investigation                 NA
#>   population_age_max population_sample_size population_country
#> 1                 NA                     NA            Nigeria
#>   population_location population_study_start_day population_study_start_month
#> 1                <NA>                         NA                         <NA>
#>   population_study_start_year population_study_end_day
#> 1                        2016                       NA
#>   population_study_end_month population_study_end_year genome_site
#> 1                       <NA>                      2018        <NA>
#>   genomic_sequence_available other_delay_start other_delay_end inverse_param
#> 1                      FALSE              <NA>            <NA>         FALSE
#>   parameter_from_figure r_pathway parameter_class parameter_type_short
#> 1                 FALSE      <NA>     Human delay                 <NA>
#>   first_author_surname year_publication     article_label
#> 1         Akhmetzhanov             2019 Akhmetzhanov 2019

^{Created on 2024-06-14 with reprex v2.1.0}

Ebola example

library(epireview)
#> Loading required package: epitrix
#> Loading required package: ggplot2
#> Loading required package: ggforce

library(epiparameter)
library(dplyr)
#> 
#> Attaching package: 'dplyr'
#> The following objects are masked from 'package:stats':
#> 
#>     filter, lag
#> The following objects are masked from 'package:base':
#> 
#>     intersect, setdiff, setequal, union

# Load Ebola data
data_in <- load_epidata(pathogen = "ebola")
#> Warning: One or more parsing issues, call `problems()` on your data frame for details,
#> e.g.:
#>   dat <- vroom(...)
#>   problems(dat)
#> Warning in load_epidata_raw(pathogen, "outbreak"): No data found for ebola
#> Warning: One or more parsing issues, call `problems()` on your data frame for details,
#> e.g.:
#>   dat <- vroom(...)
#>   problems(dat)
#> Warning in load_epidata(pathogen = "ebola"): No outbreaks information found for
#> ebola
#> Data loaded for ebola

# Extract parameters for onset to death, with qualtiy score > 50
params_in <- data_in[["params"]] 
params_infectious <- params_in |> dplyr::filter(parameter_type=="Human delay - infectious period") 

# Select distribution
params_inf_with_dist <- params_infectious |> dplyr::filter(!is.na(distribution_type))

# Start by extracting only those with Exponential
infectious <- as_epidist(params_inf_with_dist)
#> Using Lau (2017). "<title not available>." _<journal not available>_. 
#> To retrieve the citation use the 'get_citation' function
#> Warning: Cannot create full citation for epidemiological parameters without bibliographic information 
#>  see ?as_epidist for help.

unclass(infectious)
#> $disease
#> [1] "Ebola Virus Disease"
#> 
#> $pathogen
#> [1] "Ebola virus"
#> 
#> $epi_dist
#> [1] "Human delay - infectious period"
#> 
#> $prob_dist
#> <distribution[1]>
#> [1] Exp(0.25)
#> 
#> $uncertainty
#> $uncertainty$mean
#> $uncertainty$mean$ci_limits
#> [1] NA
#> 
#> $uncertainty$mean$ci
#> [1] NA NA
#> 
#> $uncertainty$mean$ci_type
#> [1] NA
#> 
#> 
#> 
#> $summary_stats
#> $summary_stats$mean
#> [1] 4.05
#> 
#> $summary_stats$mean_ci_limits
#> [1] 3.53 4.67
#> 
#> $summary_stats$mean_ci
#> [1] 95
#> 
#> $summary_stats$sd
#> [1] NA
#> 
#> $summary_stats$sd_ci_limits
#> [1] NA NA
#> 
#> $summary_stats$sd_ci
#> [1] NA
#> 
#> $summary_stats$median
#> [1] NA
#> 
#> $summary_stats$median_ci_limits
#> [1] NA NA
#> 
#> $summary_stats$median_ci
#> [1] NA
#> 
#> $summary_stats$quantiles
#> [1] NA
#> 
#> $summary_stats$range
#> [1] NA NA
#> 
#> 
#> $citation
#> Lau (2017). "<title not available>." _<journal not available>_.
#> 
#> $metadata
#> $metadata$sample_size
#> [1] 200
#> 
#> $metadata$region
#> [1] "Western Area, Sierra Leone"
#> 
#> $metadata$transmission_mode
#> [1] NA
#> 
#> $metadata$vector
#> [1] NA
#> 
#> $metadata$extrinsic
#> [1] FALSE
#> 
#> $metadata$inference_method
#> [1] "Maximum likelihood"
#> 
#> 
#> $method_assess
#> $method_assess$censored
#> [1] NA
#> 
#> $method_assess$right_truncated
#> [1] NA
#> 
#> $method_assess$phase_bias_adjusted
#> [1] NA
#> 
#> 
#> $notes
#> [1] "No additional notes"
#> 
#> attr(,".epiparameter_namespace")
#> function () 
#> NULL
#> <bytecode: 0x1211b36d0>
#> <environment: namespace:epiparameter>

as.data.frame(params_inf_with_dist)
#>                                 id                parameter_data_id
#> 1 1fb060c117ff954fa69e77429a0f9c7e a92dfc5da1285b856a016e71ce2bfbad
#>   covidence_id    pathogen                  parameter_type parameter_value
#> 1         3470 Ebola virus Human delay - infectious period            4.05
#>   exponent parameter_unit parameter_lower_bound parameter_upper_bound
#> 1        0           Days                    NA                    NA
#>   parameter_value_type parameter_uncertainty_single_value
#> 1                 Mean                                 NA
#>   parameter_uncertainty_singe_type parameter_uncertainty_lower_value
#> 1                             <NA>                              3.53
#>   parameter_uncertainty_upper_value parameter_uncertainty_type
#> 1                              4.67                     95% CI
#>   cfr_ifr_numerator cfr_ifr_denominator distribution_type
#> 1                NA                  NA       Exponential
#>   distribution_par1_value distribution_par1_type distribution_par1_uncertainty
#> 1                    4.05                   Mean                         FALSE
#>   distribution_par2_value distribution_par2_type distribution_par2_uncertainty
#> 1                      NA                   <NA>                         FALSE
#>   method_from_supplement method_moment_value cfr_ifr_method method_r
#> 1                   TRUE                <NA>           <NA>     <NA>
#>   method_disaggregated_by method_disaggregated method_disaggregated_only
#> 1                    <NA>                FALSE                     FALSE
#>   riskfactor_outcome riskfactor_name riskfactor_occupation
#> 1               <NA>            <NA>                  <NA>
#>   riskfactor_significant riskfactor_adjusted population_sex
#> 1                   <NA>                <NA>           Both
#>   population_sample_type   population_group population_age_min
#> 1       Population based General population                 NA
#>   population_age_max population_sample_size population_country
#> 1                 NA                    200       Sierra Leone
#>   population_location population_study_start_day population_study_start_month
#> 1        Western Area                         20                          Oct
#>   population_study_start_year population_study_end_day
#> 1                        2014                       30
#>   population_study_end_month population_study_end_year genome_site
#> 1                        Mar                      2015        <NA>
#>   genomic_sequence_available other_delay_start other_delay_end inverse_param
#> 1                      FALSE              <NA>            <NA>         FALSE
#>   parameter_from_figure parameter_class ebola_variant other_delay
#> 1                 FALSE     Human delay   Unspecified        <NA>
#>         delay_short       delay_start other_rf_outcome attack_rate_type
#> 1 Infectious period Infection process             <NA>             <NA>
#>   survey_start_date survey_end_date               survey_date parameter_bounds
#> 1       20 Oct 2014     30 Mar 2015 20 Oct 2014 - 30 Mar 2015             <NA>
#>   comb_uncertainty_type comb_uncertainty article_qa_score              outbreak
#> 1                95% CI      3.53 - 4.67         85.71429 West Africa 2013-2016
#>   ebola_species parameter_type_short first_author_surname year_publication
#> 1         Zaire                 <NA>                  Lau             2017
#>   article_label
#> 1  Lau 2017 (a)

^{Created on 2024-06-14 with reprex v2.1.0}

[joshwlambert]

@jfunction thanks for mentioning the replicated mean, it helped me to debug and fix the issue.

[joshwlambert]

PR #334 is now merged, closing this issue. If you have any other issues using as_epidist() or other functions in the package please feel free to open a new issue.