pigauto

Imputes missing species trait data for comparative analyses by combining three sources of information: phylogenetic similarity (closely related species share similar traits), cross-trait correlations (observed traits inform missing ones), and optional environmental covariates (climate, habitat, geography). Handles continuous measurements, counts, binary variables, ordered categories, unordered categories, bounded proportions, zero-inflated counts, and compositional multi-proportion data in a single call. The default method is a phylogenetic baseline calibrated on held-out data; an optional graph neural network correction, which is needed to use covariates, is blended in through a per-trait gate that retains a baseline fallback. Provides conformal prediction intervals for continuous, count, ordinal, and proportion traits and an experimental analysis-aware multiple-imputation workflow for one missing continuous covariate in Gaussian linear, binomial-logit, and Gaussian random-intercept models, with Rubin pooling limited to fixed effects. Stochastic graph-network and posterior-tree completions are prediction diagnostics rather than validated inferential imputations. Bundled datasets include 300-species and 9,993-species bird-trait subsets with matching example phylogenetic trees. Rubin (1987, ISBN:978-0-471-08705-2); Vovk et al. (2005, ISBN:978-0-387-25061-8); Nakagawa and de Villemereuil (2019) <doi:10.1093/sysbio/syy089>.

Releases

2

Needs compilation

No

Needs R

>= 4.1

Config/roxygen2/version

8.0.0

Config/testthat/edition

3

LazyData

true

• Phylogenetic trees and evolution

Since March 2026

• ape
• ggplot2
• graphics
• grDevices
• Matrix
• rlang
• stats
• torch
• utils
• withr

Installing it adds

32 packages, 21.3 MB of source

Of those, need compilation

17

• brms
• glmmTMB
• jomo
• jsonlite
• knitr
• lme4
• MCMCglmm
• nlme
• phangorn
• phylolm
• phytools
• pkgdown
• rgbif
• rmarkdown
• Rphylopars
• RSpectra
• smcfcs
• terra
• testthat

Lines by kind (54,202)

Lines by language

Languages are counted from file extensions.

Files

237

R functions

331

Object systems

S3: 23 methods

Median

5

Mean

9.73

Highest

151

Median lines per function

25

R functions

331

Calls

393

Unconnected

42

Groups

8

Lines counted

12,392

Lines the tests ran

8,328

Tests passed

No

Measured with

covr 3.6.5 on R 4.6.1

Has tests

Yes

Test frameworks

testthat, tinytest

Test cases

825

Lines of tests

21,191

Test lines per line of code

0.81

• Common pitfalls / FAQ
• Getting started with pigauto
• GNN architecture and the math behind pigauto
• Mixed-Type Trait Imputation
• Multiple imputation and downstream models
• Posterior-Tree Prediction Sensitivity

README

Yes, 991 words

Release notes file

Yes

Help pages

48

Lines of documentation

4,913

Commits

1,255

Open pull requests

1

Tagged releases

6

Latest tagged release

v0.10.0, 11 Jul 2026

Issues closed in

3 days (median)

Pull requests closed in

0 days (median)

Pull requests merged

98%

Created

10 May 2025

Licence file reads as

–

Method, repository practices

Across itchyshin

Example portability

The plot.pigauto_fit() example now follows the package default and fits without

the optional GNN runtime, while still demonstrating the gate and conformal plots.

The compare_methods() example now checks that torch and its Lantern runtime

can create a tensor before fitting the GNN. It reports a skip when the runtime

is unavailable, while compare_methods() keeps its documented

gnn = TRUE default.

BirdTree example and retrieval guidance

The README and tutorials now identify tree300 as a posterior sample tree and

describe how to obtain and cite BirdTree trees. The README clarifies that

BirdTree distributes Newick files and read_tree() also accepts NEXUS files.

The installed NOTICE

distinguishes the optional megatrees package licence from rights to the

underlying data, and the bundled trees300 help documents its Ericson/Hackett

mixture. The retired simulation study remains in the repository's history but

is excluded from the public site, sitemap, and search index.

The package description no longer makes an unqualified claim that the current

release has been tested at 10,000 species. The historical scaling note below

now distinguishes the completed 9,993-species missingness study from the

v0.9.0 timing run, which stopped at 5,000 tips.

New defaults: discrete traits get their own λ; safety gate and floor off

Two defaults change, and results change for anyone using the defaults.

First, binary, ordinal and categorical traits now get their own Pagel's

lambda in the baseline (on the liability scale; categorical traits through

their one-vs-rest fits). Before, categorical traits were held at lambda = 1 and

binary and ordinal traits at 1, or at the joint lambda shared with the

continuous traits on the "exact" route. The new argument

discrete_lambda = c("estimate", "fixed_1") (on impute(), fit_pigauto()

and fit_baseline(), and through ... on multi_impute() and

multi_impute_trees()) defaults to "estimate" and is recorded in the fit's

model_config. Binary and ordinal lambdas appear in $lambda_per_trait; the

categorical one-vs-rest lambdas do not. The zero-inflated-count gate column

also gets its own lambda.

Second, safety_floor and phylo_signal_gate now default to FALSE in

impute() and fit_pigauto(). With lambda estimated, the gate and the floor

did not improve accuracy in any setting tested: at Pagel's lambda of 0 to 0.1

the old defaults were worse than predicting the mean (continuous zRMSE 1.01

against 0.89 to 0.91 with them off), and on the BIEN plant data (4,745

species) that the floor was built for, turning them off changed error by

-0.002 (SE 0.001). Fits

are also 3 to 6 times faster at n = 1,000.

To restore the previous behaviour, pass

discrete_lambda = "fixed_1", safety_floor = TRUE, phylo_signal_gate = TRUE.

Evidence (docs/dev-log/discrete-lambda/README.md; simulated Brownian-motion

and OU data, n = 100 to 1,000, plus AVONET300 and BIEN): estimating discrete

lambda alone added about 0.01 to 0.04 discrete accuracy at Pagel's lambda of 0.3 or

below; together with the gate and floor off the gain was 0.04 to 0.12, and on

the four-arm study's datasets binary and categorical accuracy matched or beat

BACE's. Costs: with lambda near 1 and n of 1,000 or more, accuracy -0.002 to

-0.005 and Brier score +0.010 to +0.016; on AVONET300's two categorical

traits, Brier +0.014 to +0.020 with accuracy unchanged or better. Ordinal

accuracy still trails BACE's at Pagel's lambda of 0.3 (0.38 against 0.44 at

n = 300).

Documentation: log_transform and traits that are already logged

?impute, ?multi_impute, ?multi_impute_trees, ?preprocess_traits and the

getting-started vignette now say that the default log_transform = TRUE logs

every all-positive continuous trait, so traits that are already on a log scale

are logged twice. Set log_transform = FALSE for such data. In a real-data

check on PanTHERIA (4,027 species, four already-logged traits, posterior

draws), the default shifted Rubin-pooled regression slopes by up to 3.9

The notes are cut short here.