Skip to contents

Build custom GetOrganelle and/or MitoFinder reference databases

Source: R/custom_assembly_db.R
custom_assembly_db.Rd

Downloads mitochondrial records from NCBI GenBank for a given clade and builds reference databases for the MitoPilot Assemble module. Replaces the external Entrez Direct / python / biopython workflow with a pure-R implementation, so no external command-line tools are required.

Usage

custom_assembly_db(
  clade,
  db_path,
  db_type = c("both", "getorganelle", "mitofinder"),
  refseq_only = FALSE,
  mito_only = TRUE,
  genomic_only = TRUE,
  search_terms = NULL,
  include_nogene = TRUE,
  nogene_min_frac = 0.8,
  nogene_min_length = 12000,
  drop_no_product = TRUE,
  retain_genbank = FALSE,
  overwrite = FALSE,
  resume = FALSE,
  api_key = NULL
)

Arguments

clade

Clade (taxon) name, validated against NCBI taxonomy.

db_path

Directory in which to store the databases. Use a location outside your MitoPilot project directories so databases can be reused across projects.

db_type

Which database(s) to build: "both" (default), "getorganelle", or "mitofinder".

refseq_only

If TRUE, restrict to RefSeq mitogenomes. Default FALSE (all mitogenomes).

mito_only

If TRUE (default), restrict to records flagged as mitochondrial in origin (NCBI `mitochondrion[filter]`). Excludes nuclear records that merely mention "mitochondrial" (e.g. nuclear-encoded mito genes, NUMTs). Set FALSE to broaden recall.

genomic_only

If TRUE (default), restrict to genomic DNA (NCBI `biomol_genomic[PROP]`), excluding mRNA/rRNA/tRNA/cRNA records. Set FALSE to broaden recall.

search_terms

Optional additional NCBI advanced-query term(s) combined with the clade via AND, e.g. '"PRJNA720393"[BioProject]'. Validated before download.

include_nogene

If TRUE (default), add unannotated mitochondrial sequences to the GetOrganelle seed database when long enough (see nogene_min_frac).

nogene_min_frac

Keep an unannotated sequence if its length is at least this fraction of the median complete-mitogenome length. Default 0.8.

nogene_min_length

Absolute minimum length (bp) used when no complete mitogenomes are available to derive a threshold. Default 12000.

drop_no_product

If TRUE (default), drop single-gene sequences lacking a GenBank product (often poorly annotated) from the label database.

retain_genbank

If TRUE, keep the raw genbank.gb file for GetOrganelle-only builds. Ignored when a MitoFinder database is built (the .gb file is the MitoFinder database).

overwrite

If TRUE, replace the dated output directory if it already exists and is non-empty. If FALSE (default), an existing non-empty output directory is an error.

resume

If TRUE, continue an interrupted download instead of starting over: an existing output directory for this clade (any date) is reused and only the records not yet present in its genbank.gb are downloaded. Any partial trailing record is trimmed first. Best used soon after an interruption, while the NCBI result set is unchanged. Default FALSE. Ignored (treated as a fresh run) when no prior download exists.

api_key

Optional NCBI API key (raises the rate limit). Defaults to the NCBI_API_KEY or ENTREZ_KEY environment variable if set.

Details

Two database types are supported:

  • GetOrganelle - a seed FASTA (whole/partial mitogenomes) and a label FASTA (individual gene sequences).

  • MitoFinder - a single GenBank (.gb) file containing only complete mitochondrial genomes.

Databases are written to a dated, clade-named sub-directory of db_path (e.g. Percidae_refseq_2026-06-11), recording when GenBank was accessed. A README.txt and manifest.json capture the full query, taxid, record counts, and access time.

The historical manual review of unannotated ("nogene") sequences is automated: such sequences are added to the seed database only if they are long enough relative to the complete mitogenomes in the download (see include_nogene / nogene_min_frac).