AnVIL PopUp Workshop week 2 script

week-2-demo.R

if (!"AnVIL" %in% rownames(installed.packages()))
    BiocManager::install("AnVIL")

library(AnVIL)

this_workspace = avworkspace()
hca = "featured-workspaces-hca/HCA_Optimus_Pipeline"
thousand_genomes = "anvil-datastorage/1000G-high-coverage-2019"

## AnVIL 'namespace' / 'name' == workspace
avworkspace()
avworkspace(hca)

## DATA TABLE access
avtables()

tbl = avtable("sample")
tbl

tbl %>% count(participant)

## tbl %>% avtable_import()

## second example
avworkspace(thousand_genomes)
avtables()
participant = avtable("participant")
participant

participant %>%
    count(POPULATION, sort = TRUE)
avtable("pedigree") %>%
    count(Population, Sex) %>%
    tidyr::pivot_wider(names_from = "Sex", values_from = "n")

## switch back to the HCA workspace
avworkspace(hca)

## Google buckets

## Copy files from google buckets to persistent disk

tbl = avtable("sample_set")
tbl

dir.create("~/loom")
gsutil_cp(tbl$loom_output_file, "~/loom/")  # see also gsutil_rsync()
dir("~/loom")

## workspace bucket -- 'backup' or share persistent disk to workspace bucket

avworkspace(this_workspace)
avbucket()  # bucket associated with this workspace
gsutil_ls(avbucket())

avfiles_backup("~/scripts", recursive = TRUE) # see also avfiles_restore()
gsutil_ls(avbucket(), recursive = TRUE)

if (FALSE) {
    ## update to 'devel' version of the AnVIL package
    BiocManager::install("Bioconductor/AnVIL")  # do NOT update
                                                # out-of-date packages

    ## RESTART R, and then...
    AnVIL::repositories() # binary Bioconductor and CRAN installations

    ## install and use LoomExperiment
    AnVIL::install("LoomExperiment") # about 40 seconds, rather than
                                     # 10's of minutes
    sce = LoomExperiment::import("~/loom/pbmc_human_v3.loom")

}

## advanced -- access AnVIL from outside AnVIL, with google cloud SDK installed
## gcloud_account()
## gcloud_project()

## advanced -- the AnVIL / Terra API
leo = Leonardo()
leo
leo$listDisks()

terra = Terra()
tags(terra, "Workspaces")
wkspc = 
    terra$listWorkspaces() %>%
    flatten() %>%
    select(-starts_with("workspace.attributes"))
wkspc