Skip to contents

Changelog

Source: NEWS.md

superspreading (development version)

The third minor release of the superspreading package contains enhancements to several functions and a new vignette.

We are also pleased to welcome Dillon Adam (@dcadam) as a new package author for his contributions towards this version.

New features

  • The proportion_transmission() function has been expanded to incorporate a new method. The new method calculates the proportion of transmission from X% the most infectious individuals, corresponding to the Lloyd-Smith et al. (2005) calculation. The proportion_transmission() has a new method argument to toggle between the two calculations (@dcadam, #99).
  • A new vignette explaining the methods in the proportion_transmission() function (#101).
  • {bpmodels} is removed as a package dependency and a branching process simulation function, modified from bpmodels::chain_sim(), is added to the {superspreading} package (#103). @sbfnk is added as copyright holder and @jamesmbaazam as a contributor.
  • probability_contain() is enhanced to enable to probability of containment within a certain amount of time since the outbreak started. This adds outbreak_time and generation_time arguments to probability_contain(). This addition is backwards compatible as by default the time is unlimited, outbreak_time = Inf, and no generation time is required.

Breaking changes

Minor changes

  • The proportions output of proportion_*() functions are now formatted to significant figures rather than rounding to prevent small values being rounded to zero (#102).
  • Improve input checking, error messages and edge case handling for functions (#102).
  • Vignettes now use rmarkdown::html_vignette instead of bookdown::html_vignette2 and as_is: true has been removed due to changes to {pkgdown} in v2.1.0. {bookdown} has been removed as a suggested package and code folding is removed from vignettes. KaTeX headers have been added to _pkgdown.yml for correct math rendering (#104 & #109).
  • The get_epidist_params() internal function has been renamed get_epiparameter_params() since {epiparameter} renamed the <epidist> class to <epiparameter> (#100).
  • Internal constants have been added to the package and used by functions (#111).
  • Checking if the user specifies individual parameters (R and k) or provides an <epiparameter> object is now in .check_input_params() (#111).

Bug fixes

  • None to {superspreading} functions.
  • Update {epiparameter} use in vignette and tests (#106).

Deprecated and defunct

  • None

superspreading 0.2.0

Second minor release of superspreading. This release enhances functions added in v0.1.0 and adds two new exported functions, and two new vignettes.

New features

  • A new function (calc_network_R()) to estimate the reproduction number for heterogeneous networks and a vignette outlining use cases for the function from existing epidemiological literature is added (#71).
  • probability_epidemic() and probability_extinct() now have individual-level and population-level control in a joint framework (#70).
  • proportion_*() functions can now return proportion columns of the output <data.frame> as numeric when the new argument format_prop is set to FALSE (#72).
  • A new design principles vignette to document package development is added (#68).
  • Added a helper function (ic_tbl()) to improve model comparison tables (#65).
  • probability_*() functions now accept dynamic dots ({rlang} is added as a dependency) (#82).

Breaking changes

  • ind_control and pop_control arguments replace control and control_type arguments in probability_contain(); and the argument default for num_init_infect is removed (#70).
  • Changed epidist argument to offspring_dist to follow Epiverse style (affects several functions) (#64).
  • Argument in proportion_transmission() has been renamed from sim to simulate (#81).

Minor changes

  • New package hex logo (#73).
  • Updated continuous integration and package infrastructure (#67).
  • Improved function documentation (#63).
  • Optimisation now uses optim() by default (#82).
  • Testing suite now uses snapshot testing for regression tests (#84).

Bug fixes

  • None to {superspreading} functions.
  • Update {epiparameter} use in vignette and tests (#62).

Deprecated and defunct

  • None

superspreading 0.1.0

Initial release of superspreading, an R package to estimate individual-level variation in disease transmission and provide summary metrics for superspreading events.

New features

  • Offspring distributions, not available in base R, to fit to transmission data.
  • Functions to calculate the probability an infectious disease will cause an epidemic, go extinct or be contained.
  • Summary metric functions to calculate proportion of cases that cause a certain proportion of secondary transmission, as well as which proportion of cases are within clusters of a certain size.
  • Three vignettes, including: an introduction to the package, estimating individual-level transmission from data, and the effect of superspreading on epidemic risk.
  • Unit tests and documentation files.
  • Continuous integration workflows for R package checks, rendering the README.md, calculating test coverage, and deploying the pkgdown website.

Breaking changes

  • None

Bug fixes

  • None

Deprecated and defunct

  • None