Automatic Barcode Gap Discovery (ABGD)

Delimits species using the original algorithm of Puillandre et al. (2012). All computation (distances, gap detection, partitioning and recursion) is performed by the original C code of G. Achaz via .Call().

Usage

abgd(
  x,
  model = "simple",
  prior_min = 0.001,
  prior_max = 0.1,
  prior_steps = 10L,
  min_slope_increase = 1.5,
  ts_tv = 2
)

Arguments

x

Path to an aligned FASTA file, or an aligned DNAbin (ape) object. Distance matrices are not accepted; distances are calculated internally from the sequences. A DNAbin object is exported to a temporary FASTA file and processed by the original C code.

model

Distance model:

"simple"

p-distance with Laplace correction – equivalent to -d 3 in the original ABGD. Use this to replicate abgd -a -d 3.

"JC69"

Jukes-Cantor (original default, -d 1).

"K80"

Kimura 2-parameter (-d 0).

"TN93"

Tamura-Nei (-d 2, not implemented).

prior_min

Minimum prior for intraspecific divergence (-p). Default: 0.001.

prior_max

Maximum prior for intraspecific divergence (-P). Default: 0.1.

prior_steps

Number of steps in \([p,P]\) (-n). Default: 10.

min_slope_increase

Minimum slope increase factor (-X). Default: 1.5.

ts_tv

Transition/transversion ratio for K80 (-t). Default: 2.0.

Value

A named list with the following elements:

initial

A data.frame with one row per prior (initial partition): prior, n_groups, partition.

recursive

A data.frame with one row per prior (recursive partition): prior, n_groups, partition.

dist_matrix

Numeric distance matrix (nseq x nseq).

taxa

Character vector of sequence names.

AI Disclaimer

This function was written with assistance of AI coding agent (Claude Code Sonnet 4.6). Correctness was validated by comparing against original software, on a suite of real sequencing datasets, and manual code review. All validation and output verification was made by the authors.

Source Code

N. Puillandre, A. Lambert, S. Brouillet, G. Achaz (ABGD C engine)

References

Puillandre N, Lambert A, Brouillet S, Achaz G (2012). ABGD, Automatic Barcode Gap Discovery for primary species delimitation. Molecular Ecology, 21(8), 1864–1877. doi:10.1111/j.1365-294X.2011.05239.x

Author

Pedro S. Bittencourt

Examples

if (FALSE) { # \dontrun{
# Equivalent to: abgd -a -d 3 geophagus.fasta
result <- abgd("geophagus.fasta", model = "simple")

result$initial[, c("prior", "n_groups")]
result$recursive[, c("prior", "n_groups")]
result$recursive$partition[[1]]   # partition for the first prior

# Input from a DNAbin object
library(ape)
seqs   <- read.dna("geophagus.fasta", format = "fasta")
result <- abgd(seqs, model = "simple")
} # }