Changelog

Source: NEWS.md

eggla 1.0.4

Documentation

• In README.Rmd,
  • docs: replace previous instructions with instructions using “lock” file for dependencies.
• In vignettes/,
  • docs: remove outdated email address.

eggla 1.0.3

Fixes

• In R/*_aucs.R, R/*_slopes.R,
  • fix: sort rows by IDs (#127, thanks @DrGengWang).
  • fix: ensure denominator’s length depends on the number of knots. (#127, thanks @DrGengWang)

Tests

• In tests/testthat,
  • test: update snapshots

eggla 1.0.2

Fixes

• In R/*_aucs.R, R/*_slopes.R,
  • fix: use ID as character and use merge to combine data.frames/matrices.

Chores

• In R, tests/testthat,
  • chore: replace 1: with seq_len and $ with [[.

Tests

• In tests/testthat,
  • test: cleanup duplicated and outdated tests.
  • test: add snapshot tests.

eggla 1.0.1

Build

• In DESCRIPTION,
  • build: add growthcleanr GitHub repository as remote.

Full Changelog: https://github.com/mcanouil/eggla/compare/v1.0.0…v1.0.1

eggla 1.0.0

Build

• In inst/bin,
  • build: update PLINK2 to 5th of January 2024 version (Linux 64-bit build).
• In .github,
  • build: upgrade BCFtools to 1.19.
  • build: upgrade R to 3.4.2.
• In DESCRIPTION,
  • build: upgrade R packages to latest versions.

Chores

• In inst/CITATION,
  • chore: convert CFF to bibentry.
• In citation.cff,
  • chore: update citation file.
• In R/gsp.R,
  • chore: add @export for S3 method.

Full Changelog: https://github.com/mcanouil/eggla/compare/v0.20.0…v1.0.0

eggla 0.20.0

Fixes

• In R, and vignettes/articles,
  • fix: add missing knot at 0.75 in linear splines model.

Chores

• In .github,
  • chore: update infrastructure based on Devcontainer.

Full Changelog: https://github.com/mcanouil/eggla/compare/v0.19.6...v0.19.7

eggla 0.19.6

Fixes

• In R/run_eggla_gwas.R,
  • fix: GWAS results are now exported as a tab-delimited compressed file (i.e., txt.gz).

Full Changelog: https://github.com/mcanouil/eggla/compare/v0.19.5...v0.19.6

eggla 0.19.5

Fixes

• In R/egg_model.R,
  • fix: turns warnings into errors when fitting the model for the first time.

Full Changelog: https://github.com/mcanouil/eggla/compare/v0.19.4...v0.19.5

eggla 0.19.4

Fixes

• In R/on_attach.R, R/plot_aucs.R, R/plot_egg_aucs.R, R/predict_bmi.R, R/run_eggla_gwas.R, R/run_eggla_lmm.R,
  • fix: linters warnings.
• In R/egg_model.R,
  • fix: in case or error, also increase the umber of EM iterations.

Full Changelog: https://github.com/mcanouil/eggla/compare/v0.19.3...v0.19.4

eggla 0.19.3

Fixes

• In R/run_eggla_gwas.R,
  • fix: update commands to get version.
  • fix: ‘ax’ replaced with ‘omitted’ in PLINK2 output.

Full Changelog: https://github.com/mcanouil/eggla/compare/v0.19.2...v0.19.3

eggla 0.19.2

Build

• In inst/bin,
  • build: update PLINK2 to 11th of April 2023 version (Linux 64-bit build).

Chores

• In inst/CITATION,
  • chore: convert to bibentry.

Full Changelog: https://github.com/mcanouil/eggla/compare/v0.19.1...v0.19.2

eggla 0.19.1

Chores

• In inst/CITATION, pkgdown/_pkgdown.yml, CITATION.cff, and DESCRIPTION,
  • chore: update domain

Docs

• In R/eggla-packages.R,
  • docs: update domain.
• In R/run_eggla_gwas.R,
  • docs: update missing argument.

Full Changelog: https://github.com/mcanouil/eggla/compare/v0.19.0...v0.19.1

eggla 0.19.0

Fixes

• In R/run_eggla_gwas.R,
  • fix: update path to files in example.
  • fix: check for samples in VCF files.
  • fix: leading zeros in IDs in dataset were removed.
  • fix: use PLINK2 to compute HWE P-value, N, alleles frequencies, and missing rate.
  • fix: use PLINK2 to compute MACH_R2 imputation score.
  • fix: use vep_file as the top level parameter instead of vep.
  • fix: remove --mach-r2-filter PLINK2 option.
• In R/plot_aucs.R, R/plot_egg_aucs.R, and R/run_eggla_lmm.R,
  • fix: groupOnX argument of geom_quasirandom() is deprecated as of ggbeeswarm 0.7.1.
• In R/plot_slopes.R and R/plot_egg_slopes.R,
  • fix: using size aesthetic for lines was deprecated.
• In R/run_eggla_lmm(),
  • fix: bad copy/paste. (#100)
  • fix: wrong static title for residuals plot. (#102)
  • fix: update step parameter for adiposity peak/rebound prediction. (#98)
• In R/compute_apar.R, R/compute_outliers.R, R/egg_correlations.R, R/egg_outliers.R, and R/predict_bmi.R,
  • fix: update step parameter for adiposity peak/rebound prediction. (#98)

Chores

• chore: split .devcontainer settings into a separate repositories, i.e., .vscode and .devcontainer.

Build

• build: upgrade PLINK2 and BCFtools to latest versions.
• build: update R packages to latest versions.

Full Changelog: https://github.com/mcanouil/eggla/compare/v0.18.5...v0.19.0

eggla 0.18.5

• In R/egg_correlations.R,
  • docs: fix wrong return value in documentation.

Full Changelog: https://github.com/mcanouil/eggla/compare/v0.18.4...v0.18.5

eggla 0.18.4

Chores

• In DESCRIPTION,
  • chore: update email address.

Full Changelog: https://github.com/mcanouil/eggla/compare/v0.18.3...v0.18.4

eggla 0.18.3

Fixes

• In R/run_eggla_gwas.R,
  • fix: HWE_P should be NA for imputed variants.

Full Changelog: https://github.com/mcanouil/eggla/compare/v0.18.2...v0.18.3

eggla 0.18.2

Fixes

• In R/egg_correlations.R and R/run_eggla_lmm.R,
  • fix: parameters table not properly filtered and omitting missing values.

Full Changelog: https://github.com/mcanouil/eggla/compare/v0.18.1...v0.18.2

eggla 0.18.1

Fixes

• In R/run_eggla_gwas.R,
  • fix: info_type description given to genotyped variants.

Full Changelog: https://github.com/mcanouil/eggla/compare/v0.18.0...v0.18.1

eggla 0.18.0

Features

• In R/run_eggla_lmm.R and R/egg_correlations.R,
  • feat: perform correlations between all parameters.

Full Changelog: https://github.com/mcanouil/eggla/compare/v0.17.5...v0.18.0

eggla 0.17.5

Style

• In vignettes/articles/adiposity-peak-rebound.Rmd,
  • style: remove top ticks for AP/AR.

Full Changelog: https://github.com/mcanouil/eggla/compare/v0.17.4...v0.17.5

eggla 0.17.4

Fixes

• In R/run_eggla_gwas.R,
  • fix: rename R2 to INFO if exists.
• In vignettes/articles/models-diagnostics.Rmd,
  • revert: set back knots to original values for diagnostics.
  • fix: hardcode performance metrics to avoid changes due to performance updates.
  • fix: eval to FALSE to set w/o AR1.

Docs

• In vignettes/articles/models-diagnostics.Rmd,
  • docs: new section to update and tidy perf data.

Style

• In vignettes/articles/adiposity-peak-rebound.Rmd, vignettes/articles/run-cubic-splines.Rmd, and vignettes/articles/models-diagnostics.Rmd,
  • style: homogenise all figures with the new theme.

Full Changelog: https://github.com/mcanouil/eggla/compare/v0.17.3...v0.17.4

eggla 0.17.3

Fixes

• In R/run_eggla_lmm.R,
  • fix: order derived parameter for outlier figure.

Full Changelog: https://github.com/mcanouil/eggla/compare/v0.17.2...v0.17.3

eggla 0.17.2

Fixes

• In R/compute_apar.R,
  • fix: update filter that removed all AR estimates and add a control for AR > AP.
• In vignettes/articles/adiposity-peak-rebound.Rmd,
  • fix: missplaced parenthesis.

Full Changelog: https://github.com/mcanouil/eggla/compare/v0.17.1...v0.17.2

eggla 0.17.1

Fixes

• In R/compute_outliers.R, R/egg_outliers.R, and R/run_eggla_lmm.R,
  • fix: only allow one value for outlier_method.
• In R/plot_egg_slopes.R and R/plot_slopes.R,
  • fix: uses GAM with cubic splines every time.

Style

• style: set default ggplot2 theme to a modified version of theme_minimal() and colour/fill scales to “viridis”.
• style: use Okabe Ito colour palette.

Full Changelog: https://github.com/mcanouil/eggla/compare/v0.17.0...v0.17.1

eggla 0.17.0

Features

• In R/run_eggla_lmm.R, R/compute_outliers.R, and R/egg_outliers.R,
  • feat: new arg to pass on the outlier method.
  • feat: new arg to pass on the outlier method threshold.
  • feat: now exclude flagged outliers by setting their values to NA.

Fixes

• In R/compute_outliers.R,
  • fix: replace AP greater than 2 by NA.

Tests

• In tests,
  • test: no longer uses snapshots.

Chores

• In pkgdown/_pkgdown.yml,
  • style: separate the run_eggl_*() functions from the rest.

Full Changelog: https://github.com/mcanouil/eggla/compare/v0.16.0...v0.17.0

eggla 0.16.0

Features

• In R/run_eggla_lmm.R,
  • feat: draw outlier figure for all derived parameters.

Fixes

• In R/compute_outliers.R and R/egg_outliers.R,
  • fix: id column matching across parameters now properly works.
• In R/run_eggla_gwas.R,
  • fix: CRAN checks notes.
• In R/compute_apar.R, R/plot_aucs.R, R/plot_egg_aucs.R, R/plot_egg_slopes.R, R/plot_residuals.R, R/plot_slopes.R, vignettes/articles/adiposity-peak-rebound.Rmd, vignettes/articles/model-selection.Rmd, vignettes/articles/models-diagnostics.Rmd, and vignettes/articles/run-cubic-splines.Rmd,
  • fix: standardise ggplot default theme and fix gt tables.

Full Changelog: https://github.com/mcanouil/eggla/compare/v0.15.0...v0.16.0

eggla 0.15.0

Features

• In R/compute_outliers.R and R/egg_outliers.R,
  • feat: now detects outliers for all slopes together, all AUCs together and for individual slopes, AUCs, AP (BMI and age), and AR (BMI and age).

Full Changelog: https://github.com/mcanouil/eggla/compare/v0.14.0...v0.15.0

eggla 0.14.0

Breaking changes

• refactor!: no longer rely on renv.

Features/Fixes

• In R/run_eggla_lmm.R,
  • feat: add diagnostics output to a “to-send” folder to avoid mistakenly send individual-level data.
• In R/run_eggla_gwas.R,
  • fix: now uses the filled INFO field of VCF and compute required columns for meta-analysis.

Full Changelog: https://github.com/mcanouil/eggla/compare/v0.13.1...v0.14.0

eggla 0.13.1

Chores

• In pkgdown/_pkgdown.yml,
  • chore: update template to “minty”.

Full Changelog: https://github.com/mcanouil/eggla/compare/v0.13.0...v0.13.1

eggla 0.13.0

Breaking changes

• In R/*.R,
  • fix: knots default value is now c(1, 8, 12).
• In R/run_eggla_gwas.R,
  • refactor: path argument has been renamed working_directory for consistency with run_eggla_lmm().

Features

• In R/compute_apar.R and R/predict_bmi.R,
  • feat: filter argument filter predicted values table in case of multiple measurements.
• In R/run_eggla_lmm.R,
  • feat: filter_apar argument to be passed to compute_apar() and predict_bmi().
• In R/run_eggla_gwas.R,
  • feat: now performs GWAS of BMI and age at adiposity peak and rebound in addition to slopes and AUCs.
  • feat: now has a clean argument which clean intermediary files stored in working_directory.
  • feat: new use_info argument to indicate if INFO field should be extracted (default is FALSE).

Fixes

• In R/run_eggla_lmm.R,
  • fix: now checks if IDs are unique.
  • fix: properly export BMI and age at adiposity peak and rebound.
  • fix: PLINK2 and BCFTools binary path checks.
• In R/run_eggla_gwas.R,
  • fix: now check for individuals in common between phenotypes and genotypes before processing the VCF files.
• In DESCRIPTION,
  • fix: move some packages from suggests to imports (e.g., ggtext, ggdist, etc.)
  • fix: now uses growthcleaner from CRAN.

Chores

• In .devcontainer and inst/setup,
  • refactor: improve Docker build process and add devcontainer ability.
• In pkgdown/_pkgdown.yml,
  • refactor: reorder sections to focus on the main functions first.

Full Changelog: <https://github.com/mcanouil/eggla/compare/v0.12.3...v0.13.0>

eggla 0.12.3

Breaking changes

• In DESCRIPTION,
  • fix: depends on R >= 4.2.0.

Fixes

• In R/run_eggla_gwas.R,
  • fix: BCFtools filters as parameter with non by default.
  • fix: ensure ‘id_column’ is of type character.

Chores

• In insta/vcf and data-raw/bmigrowth.R,
  • chore: add VCF files to test run_eggla_gwas().

Tests

• In inst/setup/eggla-example.R,
  • test: add run_eggla_gwas in docker build test example.
• In tests/testthat/test-run_eggla.R,
  • test: new tests for run_eggla functions.

Full Changelog: <https://github.com/mcanouil/eggla/compare/v0.12.2...v0.12.3>

eggla 0.12.2

Fixes

• In R/run_eggla_gwas.R,
  • fix: the condition has length > 1.

Full Changelog: <https://github.com/mcanouil/eggla/compare/v0.12.1...v0.12.2>

eggla 0.12.1

Fixes

• In R/run_eggla_gwas.R,
  • fix: wrong variable in “results” input check condition.

Full Changelog: <https://github.com/mcanouil/eggla/compare/v0.12.0...v0.12.1>

eggla 0.12.0

Fixes

• In R/run_eggla_lmm.R,
  • fix: improve data transformation/cleaning (#38).
  • fix: allow to output directories path and no zip archives when zip is not available (#41).
  • fix: remove dates from model object filename.
• In R/run_eggla_gwas.R,
  • fix: allow to use directories path or zip archives as results (#41).
  • fix: rename results_zip argument to results.

Features

• In R/run_eggla_lmm.R,
  • feat: add knots and period parameters (#37).
  • feat: add clean parameter to clean directories when archives are successfully created (#41).

Chores

• chore: remove version from CFF file.

Style

• style: remove break line in badges section.

Full Changelog: <https://github.com/mcanouil/eggla/compare/v0.11.2...v0.12.0>

eggla 0.11.2

Fixes

• fix: uses general Zenodo DOI.

Full Changelog: <https://github.com/mcanouil/eggla/compare/v0.11.1...v0.11.2>

eggla 0.11.1

Chores

• chore: add DOI and citation files.

Fixes

• In vinettes/eggla.Rmd,
  • fix: uses latest moving tag.

Full Changelog: <https://github.com/mcanouil/eggla/compare/v0.11.0...v0.11.1>

eggla 0.11.0

Breaking changes

• run_eggla() has been renamed to run_eggla_lmm().
• do_eggla_gwas() has been renamed to run_eggla_gwas().

Fixes

• In R/run_eggla_lmm.R,
  • fix: exclude rds from being included in archive.
• In R/run_eggla_gwas.R,
  • fix: improve and add messages.
  • fix: ensure “data” is a data.frame or a path read in a data.frame.
  • fix: allows binary path to be an actual binary path.
  • fix: now returns results files path, input parameters zip file, and text file with software versions.
  • fix: collapse vector of covariates into an atomic string to be added in the results files.

Features

• In inst/bin,
  • feat: add PLINK2 binary internally.
• In R/run_eggla_lmm.R,
  • feat: eggla model call is now written in a file.
• In R/egg_correlations() and R/compute_correlations(),
  • feat: add correlation figure/table. (#22)
• In R/egg_outliers() and R/compute_outliers(),
  • feat: add function to check. (#22)

Tests

• In tests/testhat:
  • test: add tests for plot functions.
  • test: add more tests for time_model() and egg_model().
  • test: add tests for *_correlations() and *_outliers() functions.

Full Changelog: <https://github.com/mcanouil/eggla/compare/v0.10.4...v0.11.0>

eggla 0.10.4

Fixes

• In R/do_eggla_gwas.R,
  • Fix: plink2 binary path modified outside download if statement.
  • Fix: merge not cartesian by keeping only first measures of each individuals (#20).
• In inst/setup/Dockerfile, inst/setup/pkg.lock, inst/setup/renv.lock, anddata-raw/lock.R,
  • Fix: add all data.table dependencies.
• In R/, tests/, and vignettes/,
  • fix: documentation consistency with code, i.e., replaced “AR1” with “CAR1” (#21).

Full Changelog: <https://github.com/mcanouil/eggla/compare/v0.10.3...v0.10.4>

eggla 0.10.3

Fixes

• In inst/Dockerfile,
  • Build: add PLINK2 and install all eggla’s dependencies, including suggests.
• In R/do_eggla_gwas.R,
  • Fix: no longer relies on URL for PLINK2 since only alpha dynamic URL remains.
• In vignettes/eggla.Rmd:
  • Chore: remove unused eval = FALSE chunk option.
  • Fix: remove renv code in the Docker section.
  • Fix: set properly the working directory.

Full Changelog: <https://github.com/mcanouil/eggla/compare/v0.10.2...v0.10.3>

eggla 0.10.2

Fixes

• In R/egg_model.R,
  • Fix: try without CAR1 before reducing random effect. (#18)

Full Changelog: <https://github.com/mcanouil/eggla/compare/v0.10.1...v0.10.2>

eggla 0.10.1

Fixes

• In R/egg_slopes.R, R/plot_egg_slopes.R, R/egg_aucs.R, and R/plot_egg_aucs.R,
  • Fix: hardcoded knots, now as an argument. (#16, #17)

Full Changelog: <https://github.com/mcanouil/eggla/compare/v0.10.0...v0.10.1>

eggla 0.10.0

Features

• In R/predict_bmi.R (#14),
  • Feat: new function to predict BMI from a lme model object, from egg_model() or time_model().

Improvements

• Change default period to the following time windows for predicting ‘linear’ slopes from the model:
  • 0-0.5 years (Infancy – same as previously).
  • 1.5-3.5 years (Childhood – reduced the upper bound from 5 years).
  • 6.5-10 years (Pre-pubertal – increased the lower bound from 6 years).
  • 12-17 years (adolescence – same as previously).

Fixes

• In R/compute_apar.R (#14),
  • Fix: grep instead of any/grepl.
  • Refactor: inherits params from predict_bmi().
  • Fix: uses predict_bmi() internally.
• In R/plot_slopes.R, and R/plot_egg_slopes.R,
  • Fix: grep instead of any/grepl.
  • Fix: uses GAM ‘y ~ s(x, bs = “cr”)’.
• In vignettes/articles/adiposity-peak-rebound.Rmd; vignettes/articles/models-diagnostics.Rmd, and vignettes/articles/run-cubic-splines.Rmd (#14, #15),
  • Fix: uses GAM ‘y ~ s(x, bs = “cr”)’ and predict_bmi().
• In vignettes/eggla.Rmd,
  • Fix: “-it” instead of “–detach” for Docker command.

Full Changelog: <https://github.com/mcanouil/eggla/compare/v0.9.1...v0.10.0>

eggla 0.9.1

Features

• In R/egg_model.R, R/egg_model.R, R/run_eggla.R, and R/do_eggla_gwas,
  • Feat: add quiet to silent the messages.

Full Changelog: <https://github.com/mcanouil/eggla/compare/v0.9.0...v0.9.1>

eggla 0.9.0

Features

• In inst/setup/Dockerfile,
  • Feat: add script example (inst\setup\eggla-example.R) to run on dataset included in eggla.
• In R/egg_model.R,
  • Feat: allow to specify the knots for the splines, i.e., no longer hardcoded.

Fixes

• In R/egg_model.R,
  • Fix: hardcoded “ID” for individual specification in random effect.
• In R/time_model.R,
  • Fix: hardcoded “ID” for individual specification in random effect by adding id_var arg.
• In vignettes/eggla.Rmd (renamed/moved from vignettes/articles/run-eggla.Rmd),
  • Fix: use Rscript instead of R --no-save --no-restore --quiet.

Documents

• In vignettes/eggla.Rmd,
  • Docs: complete GWAS sections.
  • Docs: add Docker section.

Full Changelog: <https://github.com/mcanouil/eggla/compare/v0.8.1...v0.9.0>

eggla 0.8.1

Features

• In inst/setup/Dockerfile,
  • Feat: Docker image built automatically “devel” from “main” and “tag” from “releases”.

Full Changelog: <https://github.com/mcanouil/eggla/compare/v0.8.0...v0.8.1>

eggla 0.8.0

Features

• In R/compute_apar.R,
  • Feat: allow to compute adiposity peak and adiposity rebound on raw data from a "lme" model object.
  • Refactor: rename arguments.
• In tests, DESCRIPTION,
  • Feat: now uses the third edition of testthat.

Bug Fixes

• In R/do_eggla_gwas.R,
  • Fix: no visible global function definition for ‘patterns’.

Build

• In inst/setup,
  • build: update lock files from renv and pak based on local package.
• In data-raw/lock.R,
  • build: update script to work on local version for both renv and pak.

Full Changelog: <https://github.com/mcanouil/eggla/compare/v0.7.0...v0.8.0>

eggla 0.7.0

Features

• In R/compute_apar.R,
  • Feat: Allows to define the range in which to compute adiposity peak and adiposity rebound.
• In R/egg_model.R, R/time_model.R,
  • Feat: Allows to use (or not) AR(1) auto-correlation in the model, if TRUE (default: FALSE), the function will try without if all models fail.

Fixes

• In vignettes/articles/run-eggla.Rmd, vignettes/articles/models-diagnostics.Rmd,
  • Fix: Uses CAR1 by default in vignettes, only to speed-up computation.

Full Changelog: <https://github.com/mcanouil/eggla/compare/v0.6.1...v0.7.0>

eggla 0.6.1

Fixes

• In R/compute_apar.R,
  • Fix: returns only the first local maxima (adiposity peak) and minima (adiposity rebound).

Full Changelog: <https://github.com/mcanouil/eggla/compare/v0.6.0...v0.6.1>

eggla 0.6.0

New Feature

• In R/do_eggla_gwas.R:
  • Feat: Function to run GWAS on eggla results using PLINK2 (and BCFtools).
• In vignettes/articles/run_eggla,
  • Feat: Add “GWAS” step.
  • Refactor: use “R EOF” syntax.

Fixes

• In R/run_eggla.R:
  • Fix: Rename (add “s”) AUCs csv output file for consistency.
  • Docs: Add missing documentation for parameters.
• In R/plot_slopes.R:
  • Fix: hardcoded “age” variable.
  • Fix: ignore case in variable search.
• In R/plot_egg_slopes.R:
  • Fix: hardcoded “age” variable.
  • Fix: ignore case in variable search.
• In R/compute_apar.R:
  • Fix: ignore case in variable search.

Full Changelog: <https://github.com/mcanouil/eggla/compare/v0.5.2...v0.6.0>

eggla 0.5.2

• Fix: renv::restore issue with renv.lock from GitHub.
• Fix: Update pak install command.
• Chore: Upgrade lock files.

Full Changelog: <https://github.com/mcanouil/eggla/compare/v0.5.1...v0.5.2>

eggla 0.5.1

• No longer uses nlme::corCAR1(form = ~ 1 | ID) in models.
• Add Kimberley Burrows (@burrowsk) and Anni Heiskala as authors.

Full Changelog: <https://github.com/mcanouil/eggla/compare/v0.5.0...v0.5.1>

eggla 0.5.0

• Improve/Polish documentation,
  • Add title.
  • Add/tweak description.
  • Add code examples.
• In R/egg_*.R,
  • Add random_complexity parameter.
  • Remove _var arguments.
• In R/compute_apar.R,
  • New function to compute adiposity peak and adiposity rebound from a lme model object, such as egg_model.
• In R/compute_aucs.R,
  • Rename function for consistency, previously known as compute_auc.
• In R/compute_slopess.R,
  • Rename function for consistency, previously known as predict_average_slopes.
• In vignettes/articles,
  • Add 01-models-diagnostics.Rmd, to describe models testing and diagnostics.
  • Add 02-models-selection.Rmd, to show how to perform models selection based on results from models-diagnostics.Rmd.
  • Add 03-run-cubic-splines.Rmd, to show how to run cubic splines models on the example dataset (previously in README.Rmd).
  • Add 04-adiposity-peak-rebound.Rmd, to identify/caracterise adiposity peak and adiposity rebound from models-diagnostics.Rmd.
  • Add 99-run-eggla.Rmd, to show how to run analysis interactively or not.
• In R/plot_*.R,
  • New functions to plot area under the curves and slopes.
• In pkgdown/_pkgdown.yml,
  • Polish website.

Full Changelog: <https://github.com/mcanouil/eggla/compare/v0.4.4...v0.5.0>

eggla 0.4.4

• In R/egg_model.R,
  • Three cublic spline random effects formula are tested in the following order:
    1. ~ gsp(age, knots = c(2, 8, 12), degree = rep(3, 4), smooth = rep(2, 3)) | ID
    2. ~ gsp(age, knots = c(2, 8, 12), degree = rep(3, 4), smooth = rep(2, 3))[,1:3] | ID
    3. ~ gsp(age, knots = c(2, 8, 12), degree = rep(1, 4), smooth = rep(2, 3)) | ID
  • Add message about which model is currently being computed.
• Add pkgdown website.

Full Changelog: <https://github.com/mcanouil/eggla/compare/v0.4.3...v0.4.4>

eggla 0.4.3

• In README.(R)md,
  • Disable renv caching.
• In inst/model-diagnostics,
  • Remove ragg dependency, uses grDevices instead.
• In inst/setup,
  • Add Dockerfile.
  • Use pak in addition to renv.
• In R/run_eggla.R,
  • Remove ragg dependency, uses grDevices instead.

Full Changelog: <https://github.com/mcanouil/eggla/compare/v0.4.2...v0.4.3>

eggla 0.4.2

• In README.(R)md,
  • Fix typo in renv::init calls.

Full Changelog: <https://github.com/mcanouil/eggla/compare/v0.4.1...v0.4.2>

eggla 0.4.1

• In README.(R)md,
  • Add renv cleaning in bash example script.
  • Add setup for both interactive and non-interactive use.
• In R/run_eggla.R,
  • Switch from svglite to ragg to avoid issues with high number of elements.
  • Fix id_var which was not set with id_var = "egg_id" internally.
• In DESCRIPTION,
  • Switch from svglite to ragg to avoid issues with high number of elements.

Full Changelog: <https://github.com/mcanouil/eggla/compare/v0.4.0...v0.4.1>

eggla 0.4.0

• In README.(R)md,
  • Add table of content.
  • Add Bash script to run everything.
• In inst/setup,
  • Add renv.lock file to list dependencies with version and help install or restore them.
• In R/egg_model.R,
  • Mixed model selected inside a helper/wrapper function.
• In R/run_eggla.R,
  • Compute quaility-control for BMI.
  • Compute modelling.
  • Write model object.
  • Write derived parameters.
  • Generate residuals plot.
• In R/time_model.R,
  • Fix model specification where additional covariates were not properly added to the model formula.
• In R/plot_residuals.R,
  • Fix hardcoded variable, i.e., “age”.
• In inst/models-diagnostics,
  • models-diagnostics.R, R script to compute different models and extract performance metrics.
  • models-best.R, R script to compare models performance from models-diagnostics.R.
  • models-adiposity.R, R script to compare predicted BMI values to check for the “adiposity peak”.
• In R,
  • Remove trailing spaces.
  • Comment code not used.
  • Add missing documentation.
• In LICENSE,
  • Add Nicole Warrington.

Full Changelog: <https://github.com/mcanouil/eggla/compare/v0.3.0...v0.4.0>

eggla 0.3.0

• In R/time_model.R,
  • Add fall back (simpler random effect) models for “cubic_splines”, “cubic_slopes” and “linear_splines”.
  • Fix typos in messages.
  • Model call is printed as message.
• In inst/rmarkdown/templates/eggla/skeleton/skeleton.Rmd,
  • Add Daymont flag exclusion argument and control before modelling step.
  • Summary table and models use covariates parameter.
  • Uses “sex” coded as 1 for male and 0 for female.
  • Small code refactoring.
  • Use performance to compare models.
  • Add summary_variables parameter to show in the summary table.
  • Decrease font size in Daymont’s QC summary table .
  • Fix numbers in the summary table.

Full Changelog: <https://github.com/mcanouil/eggla/compare/v0.2.0...v0.3.0>

eggla 0.2.0

• In R/time_model.R,
  • Fix output when using as_text = TRUE.
  • Add cov to allow additional covariates.
• In inst/rmarkdown/templates/eggla/skeleton/skeleton.Rmd,
  • Add cov to allow additional covariates.

Full Changelog: <https://github.com/mcanouil/eggla/compare/v0.1.0...v0.2.0>

eggla 0.1.0

• Initial version.