Skip to contents

Initialize a RiboPilot test project

Source: R/init_test_project.R
new_test_project.Rd

Sets up a ready-to-run test project from the bundled test set: 10 fish genome skims (USNM, study SRP313821) plus two synthetic samples, CHIM00 (a clean control) and CHIM01 (fish rDNA joined to a Penicillium 28S). Downloads each fish sample's paired reads, writes the mapping file, and initializes the project database and config.

Usage

new_test_project(
  path = ".",
  n = Inf,
  full_size = FALSE,
  executor = "docker",
  container = paste0("macguigand/ribopilot:", utils::packageVersion("RiboPilot")),
  min_depth = 500,
  sra_bin = NULL,
  Rproj = TRUE,
  force = FALSE,
  fetch_metadata = TRUE,
  ...
)

Arguments

path

Project directory (created if missing; default current dir).

n

Number of downloaded samples to include (default all 10). The bundled synthetic samples CHIM00 and CHIM01 are always included.

full_size

Fetch the complete read sets from ENA/SRA (TRUE) instead of the small rDNA-filtered subsets (default FALSE).

executor

Executor / config profile (default "docker").

container

Container image for the pipeline.

min_depth

Minimum number of reads left after trimming, counting both reads of a pair (2000000 is 1000000 pairs). A sample below it is not assembled and is marked Failed with a note.

sra_bin

Directory holding fasterq-dump (for the SRA fallback); default searches PATH.

Rproj

Initialize an RStudio project (default TRUE; no effect outside RStudio).

force

Recreate an existing project database / config (default FALSE).

fetch_metadata

Fetch each sample's NCBI metadata (synthetic samples use their source run's) and follow links to GBIF and GEOME (default TRUE). Linking is saved as the project setting. Set FALSE when offline.

...

Passed to new_project() (e.g. custom_seeds_db).

Value

(invisibly) the project path.

Details

By default the reads are the rDNA-filtered subsets published with the RiboPilot GitHub releases (a few MB per sample, seconds to fetch). With full_size = TRUE the complete read sets are fetched instead, from ENA when reachable and falling back to NCBI SRA (fasterq-dump, which must be on PATH or given via sra_bin) on networks where the ENA fastq mirror is blocked; that is several GB and takes a while. Use n to fetch fewer samples. Already-downloaded, valid read files are skipped, so an interrupted run resumes cleanly.