diff --git a/README.md b/README.md index 46e570c..af22322 100644 --- a/README.md +++ b/README.md @@ -8,7 +8,7 @@ All code is in [R](https://cran.r-project.org/). The [targets package](https://d Data files need to be downloaded from three locations. -1. Dataset on Dryad for this project: https://doi.org/doi:10.5061/dryad.w0vt4b8s2 (LINK NOT LIVE YET). Cick on the "download dataset" icon, download the zipped dataset, then unzip it and put the contents in the `data/` folder in this repo. +1. Dataset on FigShare for this project: https://doi.org/10.6084/m9.figshare.16655263 (LINK NOT LIVE YET). Cick on the "Download all" icon, download the zipped dataset, then unzip it and put the contents in the `data/` folder in this repo. 2. Dataset on Dryad for Ebihara and Nitta 2019: https://datadryad.org/stash/dataset/doi:10.5061/dryad.4362p32. Download the zipped dataset and put in the `data/` folder directly (without unzipping). 3. Dataset on FigShare for FTOL v0.0.1 (Nitta et al, in prep): https://doi.org/doi:10.6084/m9.figshare.13256801 (LINK NOT LIVE YET). Download the zipped dataset and put in the `data/` folder directly (without unzipping). @@ -50,15 +50,49 @@ When you're done, take down the container: docker-compose down ``` -## Using the `targets` cache +## Targets cache -The analysis includes some steps that take a long time to run, especially maximum-likelihood phylogenetic analysis (ca. 1 week with 10 cores in parallel). To avoid running the entire workflow from scratch, untar the `_targets.tar.gz` file in the Dryad dataset and place it in the root of this repo as `_targets`: +The [targets package](https://docs.ropensci.org/targets/index.html) manages the workflow and saves all intermediate analysis results to a folder named `_targets`; this is the targets cache. +Normally, you would have to run all of the analyses starting from the original data files to generate all of the analysis results, as described above. +This takes a long time. The longest step is the phylogenetic analysis, which takes about 1 week using 10 cores in parallel. + +I have put the targets cache for this project [on github](https://github.com/joelnitta/japan_ferns_spatial_phy_cache) (LINK NOT LIVE YET) under version control using the [gittargets package](https://github.com/ropensci/gittargets). + +So instead of running everything from scratch, you can checkout the exact results matching a specific code version as follows (this assumes we are in the `japan_ferns_spatial_phy` folder and requires git): + +1. Clone the targets cache to a folder called `_targets`. + +``` +git clone https://github.com/joelnitta/japan_ferns_spatial_phy_cache _targets +``` + +2. Enter the `_targets` directory. + +``` +cd _targets +``` + +3. Fetch branches from the remote repo ([each branch corresponds to a selected commit in the code](https://docs.ropensci.org/gittargets/articles/git.html#snapshot-model)). ``` -tar -xzf _targets.tar.gz +git fetch ``` -Then, when you open the project in R [as described above](#interacting-with-the-code), you can use `targets::tar_load()` to load any target (intermediate workflow step) listed in [`_targets.R`](_targets.R). For more information on how to use the `targets` package, see https://github.com/ropensci/targets. +4. Change to the latest branch (the part of the name after `code=` matches the corresponding commit hash in `japan_ferns_spatial_phy`). + +``` +git switch code=0f9744508fbdc1d22319faa6118c2811c34c0c7d +``` + +5. Move back up to the `japan_ferns_spatial_phy` folder. + +``` +cd .. +``` + +You can also change between different snapshots of the targets cache and code using [gittargets](https://github.com/ropensci/gittargets). + +When you open the project in R [as described above](#interacting-with-the-code), you can use `targets::tar_load()` to load any target (intermediate workflow step) listed in [`_targets.R`](_targets.R). For more information on how to use the `targets` package, see https://github.com/ropensci/targets. ## Licenses @@ -66,4 +100,3 @@ Then, when you open the project in R [as described above](#interacting-with-the- - Data: [CC0 1.0 license](https://creativecommons.org/publicdomain/zero/1.0/) - [Manuscript (preprint)](https://doi.org/10.1101/2021.08.26.457744): [CC BY-NC-ND 4.0 license](https://creativecommons.org/licenses/by-nc-nd/4.0/) - [Roboto font](https://github.com/google/roboto/): [Apache 2.0 license](http://www.apache.org/licenses/LICENSE-2.0) - diff --git a/ms/data_readme.Rmd b/ms/data_readme.Rmd index d05e879..7e20677 100644 --- a/ms/data_readme.Rmd +++ b/ms/data_readme.Rmd @@ -95,11 +95,11 @@ Licenses/restrictions placed on the data, or limitations of reuse: CC0 1.0 Universal (CC0 1.0) Recommended citation for the data: Nitta JH, Mishler BD, Iwasaki W, Ebihara A -(2021) Data from: Spatial phylogenetics of Japanese ferns: Patterns, processes, +(2022) Data from: Spatial phylogenetics of Japanese ferns: Patterns, processes, and implications for conservation FIXME: add DOI when available Citation for and links to publications that cite or use the data: Nitta JH, -Mishler BD, Iwasaki W, Ebihara A (2021) Spatial phylogenetics of Japanese ferns: +Mishler BD, Iwasaki W, Ebihara A (2022) Spatial phylogenetics of Japanese ferns: Patterns, processes, and implications for conservation FIXME: add journal when published Code for analyzing the data is available on github: @@ -114,7 +114,6 @@ DATA & FILE OVERVIEW File list (filenames, directory structure (for zipped files) and brief description of all data files): -- _targets.tar.gz: Tarball (compressed folder) including all workflow results produced by R targets package - japan_climate.gpkg: Climate data in Japan downloaded from WorldClim database - japan_deer_range.gpkg: Distribution maps of Japanese deer (Cervus nippon) in Japan - japan_ferns_comm_full.csv: Community matrix (species x sites matrix) of native, non-hybrid ferns in Japan, full (unfiltered) dataset @@ -191,7 +190,7 @@ Data files were generated from raw data (not included here) using scripts available at https://github.com/joelnitta/japan_ferns_spatial_phy, in particular https://github.com/joelnitta/japan_ferns_spatial_phy/blob/main/R/process_raw_data.R. -For full methods, see Nitta JH, Mishler BD, Iwasaki W, Ebihara A (2021) Spatial +For full methods, see Nitta JH, Mishler BD, Iwasaki W, Ebihara A (2022) Spatial phylogenetics of Japanese ferns: Patterns, processes, and implications for conservation FIXME: Add journal when published @@ -201,22 +200,6 @@ DATA-SPECIFIC INFORMATION \-\-\-\-\-\-\-\-\-\-\-\-\-\-\-\-\-\-\-\-\-\-\-\-\-\- -_targets.tar.gz: Tarball (compressed folder) including all workflow results -produced by R targets package. This is provided to enable inspection of workflow -steps without running the entire workflow from the beginning. To use it, unpack -the tar achive with the command "tar -xzf _targets.tar.gz". Then, in R, the -"tar_load()" function in the R package "targets" can be used to load any -workflow step (target) defined in _targets.R -(https://github.com/joelnitta/japan_ferns_spatial_phy/blob/main/_targets.R). For -more information on the structure of the _targets folder and how to use it, see -https://github.com/ropensci/targets. - -MD5 checksum: FIXME (add manually, since this can't be calculated from inside the targets workflow) - -Corresponding commit in repo (https://github.com/joelnitta/japan_ferns_spatial_phy): FIXME add manually - -\-\-\-\-\-\-\-\-\-\-\-\-\-\-\-\-\-\-\-\-\-\-\-\-\-\- - ```{r deer-range} deer_range <- sf::st_read(here::here(deer_range_file)) # Check that no data are missing @@ -668,11 +651,11 @@ japan_ferns_biodiv_figshare <- read_csv(here::here(japan_ferns_biodiv_figshare_f `r fs::path_file(japan_ferns_biodiv_figshare_file)` (contained in "results.zip"): Biodiversity statistics of native, non-hybrid ferns and environmental variables -in Japan. Biodiversity metrics calculated as described in Nitta et al. 2021. +in Japan. Biodiversity metrics calculated as described in Nitta et al. 2022. Climatic (temperature and preciptation) variables calculated as described for japan_climate.gpkg. Includes one row with missing environmental data and one outlier for % apomixis that were removed prior to -spatial modeling analysis in Nitta et al. 2021. +spatial modeling analysis in Nitta et al. 2022. Number of variables: `r ncol(japan_ferns_biodiv_figshare)`