add est.incidence.by test

.github/workflows/lint-changed-files.yaml

@@ -16,14 +16,16 @@ jobs:
       - uses: actions/checkout@v4
 
       - uses: r-lib/actions/setup-r@v2
+        with:
+          use-public-rspm: true
 
-      - uses: r-lib/actions/setup-r-dependencies@v2
+      - uses: r-lib/actions/setup-r-dependencies@HEAD
         with:
           extra-packages: |
             any::gh
             any::lintr
             any::purrr
-          needs: check
+          needs: lint
 
       - name: Add lintr options
         run: |

DESCRIPTION

@@ -1,7 +1,7 @@
 Type: Package
 Package: serocalculator
 Title: Estimating Infection Rates from Serological Data
-Version: 1.2.0.9020
+Version: 1.2.0.9021
 Authors@R: c(
     person("Peter", "Teunis", , "p.teunis@emory.edu", role = c("aut", "cph"),
            comment = "Author of the method and original code."),

NEWS.md

@@ -1,6 +1,9 @@
 # serocalculator (development version)
 
 ## New features
+
+* Added example datasets with documentation for examples and testing (#314)
+
 * Improved error messaging for `autoplot.pop_data()` (#234).
 
 * Clarified package installation instructions in scrub typhus vignette (#234).
@@ -25,6 +28,10 @@
 ## Internal changes
 * Change default pipe setting (#312)
 
+* Added `snapshot_value` test for `est.incidence()` (#315)
+
+* Sped up `lint-changed-files` GitHub Action (#317)
+
 * Added online preview builds for PRs that change the `pkgdown` website (#309)
 
 * Added `test-autoplot.pop_data` test (#234)

R/example_noise_params_pk.R

@@ -0,0 +1,26 @@
+#' @title Small example of noise parameters for typhoid
+#'
+#' @description A subset of noise parameter estimates from the SEES study,
+#' for examples and testing.
+#'
+#' @format ## `example_noise_params_pk`
+#' A `curve_params` object (from [as_curve_params()]) with 4 rows and 7 columns:
+#' \describe{
+#'   \item{antigen_iso}{which antigen and isotype are being measured
+#'   (data is in long format)}
+#'   \item{Country}{Location for which the noise parameters were estimated}
+#'   \item{y.low}{Lower limit of detection}
+#'   \item{eps}{Measurement noise, defined by a CV (coefficient of variation)
+#'   as the ratio of the standard deviation to the mean for replicates.
+#'   Note that the CV should ideally be measured across plates
+#'   rather than within the same plate.}
+#'   \item{nu}{Biological noise: error from
+#'   cross-reactivity to other antibodies.
+#'   It is defined as the 95th percentile of
+#'   the distribution of antibody responses to the antigen-isotype
+#'   in a population with no exposure.}
+#'   \item{y.high}{Upper limit of detection}
+#'   \item{Lab}{Lab for which noise was estimated.}
+#' }
+#' @source <https://osf.io/rtw5k>
+"example_noise_params_pk"

R/example_typhoid_curves_nostrat.R

@@ -0,0 +1,19 @@
+#' Small example of antibody response curve parameters for typhoid
+#'
+#' A subset of data from the SEES study, for examples and testing.
+#'
+#' @format ## `typhoid_curves_nostrat_100`
+#' A `curve_params` object (from [as_curve_params()]) with 500 rows and 7
+#' columns:
+#' \describe{
+#'   \item{antigen_iso}{which antigen and isotype are being measured
+#'   (data is in long format)}
+#'   \item{iter}{MCMC iteration}
+#'   \item{y0}{Antibody concentration at t = 0 (start of active infection)}
+#'   \item{y1}{Antibody concentration at t = `t1` (end of active infection)}
+#'   \item{t1}{Duration of active infection}
+#'   \item{alpha}{Antibody decay rate coefficient}
+#'   \item{r}{Antibody decay rate exponent parameter}
+#' }
+#' @source <https://osf.io/rtw5k>
+"typhoid_curves_nostrat_100"

R/sees_pop_data_pakistan_100.R

@@ -0,0 +1,37 @@
+#' Small example cross-sectional data set
+#'
+#' A subset of data from the SEES data, for examples and testing.
+#'
+#' @format ## `sees_pop_data_pk_100`
+#' A `pop_data` object (from [as_pop_data()]) with 200 rows and 8 columns:
+#' \describe{
+#'   \item{id}{Observation ID}
+#'   \item{Country}{Country where the participant was living}
+#'   \item{cluster}{survey sampling cluster}
+#'   \item{catchment}{survey catchment area}
+#'   \item{age}{participant's age when sampled, in years}
+#'   \item{antigen_iso}{which antigen and isotype are being measured
+#'   (data is in long format)}
+#'   \item{value}{concentration of antigen isotype, in ELISA units}
+#' }
+#' @source <https://osf.io/n6cp3>
+"sees_pop_data_pk_100"
+
+#' Small example cross-sectional data set
+#'
+#' A subset of data from the SEES data, for examples and testing.
+#'
+#' @format ## `sees_pop_data_pk_100_old_names`
+#' A `pop_data` object (from [as_pop_data()]) with 200 rows and 8 columns:
+#' \describe{
+#'   \item{index_id}{Observation ID}
+#'   \item{Country}{Country where the participant was living}
+#'   \item{cluster}{survey sampling cluster}
+#'   \item{catchment}{survey catchment area}
+#'   \item{Age}{participant's age when sampled, in years}
+#'   \item{antigen_iso}{which antigen and isotype are being measured
+#'   (data is in long format)}
+#'   \item{result}{concentration of antigen isotype, in ELISA units}
+#' }
+#' @source <https://osf.io/n6cp3>
+"sees_pop_data_pk_100_old_names"

data-raw/example_noise_params_pk.R

@@ -0,0 +1,5 @@
+example_noise_params_pk <-
+  load_noise_params("https://osf.io/download//hqy4v/") %>%
+  filter(Country == "Pakistan")
+
+usethis::use_data(example_noise_params_pk, overwrite = TRUE)

data-raw/sees_pop_data_pakistan_100.R

@@ -0,0 +1,23 @@
+sees_pop_data_pk_100 <- load_pop_data(
+  file_path = "https://osf.io/download//n6cp3/",
+  age = "Age",
+  value = "result",
+  id = "index_id",
+  standardize = TRUE
+) %>%
+  filter(Country == "Pakistan") %>%
+  slice_head(n = 100, by = antigen_iso)
+
+usethis::use_data(sees_pop_data_pk_100, overwrite = TRUE)
+
+sees_pop_data_pk_100_old_names <- load_pop_data(
+  file_path = "https://osf.io/download//n6cp3/",
+  age = "Age",
+  value = "result",
+  id = "index_id",
+  standardize = FALSE
+) %>%
+  filter(Country == "Pakistan") %>%
+  slice_head(n = 100, by = antigen_iso)
+
+usethis::use_data(sees_pop_data_pk_100_old_names, overwrite = TRUE)

data-raw/typhoid_curves_nostrat_100.R

@@ -0,0 +1,5 @@
+typhoid_curves_nostrat_100 <-
+  load_curve_params("https://osf.io/download/rtw5k/") %>%
+  dplyr::filter(iter %in% 1:100)
+
+usethis::use_data(typhoid_curves_nostrat_100, overwrite = TRUE)

tests/testthat/_snaps/est.incidence.md

@@ -1,11 +1,19 @@
 # est.incidence() produces expected results for typhoid data
 
     Code
-      typhoid_results
+      summary(typhoid_results)
     Output
       # A tibble: 1 x 10
         est.start incidence.rate     SE CI.lwr CI.upr coverage log.lik iterations
             <dbl>          <dbl>  <dbl>  <dbl>  <dbl>    <dbl>   <dbl>      <int>
-      1       0.1          0.133 0.0216 0.0962  0.182     0.95   -261.          4
+      1       0.1          0.166 0.0178  0.135  0.205     0.95   -524.          5
       # i 2 more variables: antigen.isos <chr>, nlm.convergence.code <ord>
 
+---
+
+    structure(list(minimum = 523.575044823023, estimate = -1.7955958453869, 
+        gradient = 3.60891331241403e-06, hessian = structure(86.991906300701, dim = c(1L, 
+        1L)), code = 1L, iterations = 5L), class = c("seroincidence", 
+    "list"), lambda_start = 0.1, antigen_isos = c("HlyE_IgG", "HlyE_IgA"
+    ))
+

tests/testthat/_snaps/log_likelihood.md

@@ -1,4 +1,4 @@
 # `log_likelihood()` gives consistent results
 
-    -9268.8238
+    -533.379886031329