metacoder (version 0.3.8)

ncbi_taxon_sample: Download representative sequences for a taxon

Description

Downloads a sample of sequences meant to evenly capture the diversity of a given taxon. Can be used to get a shallow sampling of vast groups. CAUTION: This function can make MANY queries to Genbank depending on arguments given and can take a very long time. Choose your arguments carefully to avoid long waits and needlessly stressing NCBI's servers. Use a downloaded database and a parser from the taxa package when possible.

Usage

ncbi_taxon_sample(
  name = NULL,
  id = NULL,
  target_rank,
  min_counts = NULL,
  max_counts = NULL,
  interpolate_min = TRUE,
  interpolate_max = TRUE,
  min_children = NULL,
  max_children = NULL,
  seqrange = "1:3000",
  getrelated = FALSE,
  fuzzy = TRUE,
  limit = 10,
  entrez_query = NULL,
  hypothetical = FALSE,
  verbose = TRUE
)

Arguments

name

(character of length 1) The taxon to download a sample of sequences for.

id

(character of length 1) The taxon id to download a sample of sequences for.

target_rank

(character of length 1) The finest taxonomic rank at which to sample. The finest rank at which replication occurs. Must be a finer rank than taxon.

min_counts

(named numeric) The minimum number of sequences to download for each taxonomic rank. The names correspond to taxonomic ranks.

max_counts

(named numeric) The maximum number of sequences to download for each taxonomic rank. The names correspond to taxonomic ranks.

interpolate_min

(logical) If TRUE, values supplied to min_counts and min_children will be used to infer the values of intermediate ranks not specified. Linear interpolation between values of specified ranks will be used to determine values of unspecified ranks.

interpolate_max

(logical) If TRUE, values supplied to max_counts and max_children will be used to infer the values of intermediate ranks not specified. Linear interpolation between values of specified ranks will be used to determine values of unspecified ranks.

min_children

(named numeric) The minimum number sub-taxa of taxa for a given rank must have for its sequences to be searched. The names correspond to taxonomic ranks.

max_children

(named numeric) The maximum number sub-taxa of taxa for a given rank must have for its sequences to be searched. The names correspond to taxonomic ranks.

seqrange

(character) Sequence range, as e.g., "1:1000". This is the range of sequence lengths to search for. So "1:1000" means search for sequences from 1 to 1000 characters in length.

getrelated

(logical) If TRUE, gets the longest sequences of a species in the same genus as the one searched for. If FALSE, returns nothing if no match found.

fuzzy

(logical) Whether to do fuzzy taxonomic ID search or exact search. If TRUE, we use xXarbitraryXx[porgn:__txid<ID>], but if FALSE, we use txid<ID>. Default: FALSE

limit

(numeric) Number of sequences to search for and return. Max of 10,000. If you search for 6000 records, and only 5000 are found, you will of course only get 5000 back.

entrez_query

(character; length 1) An Entrez-format query to filter results with. This is useful to search for sequences with specific characteristics. The format is the same as the one used to seach genbank. (https://www.ncbi.nlm.nih.gov/books/NBK3837/#EntrezHelp.Entrez_Searching_Options)

hypothetical

(logical; length 1) If FALSE, an attempt will be made to not return hypothetical or predicted sequences judging from accession number prefixs (XM and XR). This can result in less than the limit being returned even if there are more sequences available, since this filtering is done after searching NCBI.

verbose

(logical) If TRUE, progress messages will be printed.

Examples

# \donttest{
# Look up 5 ITS sequences from each fungal class
data <- ncbi_taxon_sample(name = "Fungi", target_rank = "class", limit = 5, 
                          entrez_query = '"internal transcribed spacer"[All Fields]')

# Look up taxonomic information for sequences
obj <- lookup_tax_data(data, type = "seq_id", column = "gi_no")

# Plot information
metacoder::filter_taxa(obj, taxon_names == "Fungi", subtaxa = TRUE) %>% 
  heat_tree(node_label = taxon_names, node_color = n_obs, node_size = n_obs)
# }