phiper

phiper is an end-to-end R toolkit for Phage ImmunoPrecipitation sequencing (PhIP-seq) data analysis. It covers everything from quality control and enrichment statistics to diversity metrics and publication-ready figures — supporting both cross-sectional and longitudinal study designs.

Installation

You can install the development version of phiper from GitHub with either pak or devtools:

# install.packages("pak")
pak::pak("Polymerase3/phiper")

# or, using devtools:
# install.packages("devtools")
devtools::install_github("Polymerase3/phiper")

Note: phiper depends on phiperio for data import and class construction. It is installed automatically by the commands above.

What can phiper do?

phiper is organised into four analysis domains, each with dedicated computation and visualisation functions:

Enrichment counts — plot_enrichment_counts() gives a quick overview of per-group peptide hit counts.
Alpha diversity — richness, Shannon, Simpson, Pielou evenness, and Berger-Parker dominance, with optional significance testing and bracket annotations (compute_alpha(), compute_alpha_significance()).
Beta diversity — pairwise distances, PCoA, CAP (constrained ordination), PERMANOVA, dispersion tests, and t-SNE (compute_distance(), compute_pcoa(), compute_capscale(), compute_permanova(), compute_dispersion(), compute_tsne()).
Global prevalence shift (delta) — subject-level permutation tests for antigen-wide prevalence shifts between groups, with forest plots and ECDF visualisations (compute_delta(), forestplot(), ecdf_plot()).
Per-feature prevalence comparison (POP) — Fisher’s exact and McNemar tests for every peptide or taxonomic rank, visualised as scatter plots and volcano plots (compute_pop(), scatter_static(), volcano_static()).

The `compute_` / `plot_` pattern

Every domain follows the same two-step workflow: a compute_*() function returns a tidy data structure, which is then passed to one or more plot_*() functions. This keeps computation and visualisation cleanly separated and makes it easy to inspect, filter, or export results before plotting.

Static and interactive outputs

All visualisation functions come in two flavours: a static version built on ggplot2 (composable with the full tidyverse graphics ecosystem) and an interactive version built on plotly (useful for exploratory data analysis and HTML reports). Both use the built-in theme_phip() and phiper colour scales for a consistent visual style.

Documentation, examples, and vignettes

Full documentation, worked examples, and vignettes are available on the package website:

https://polymerase3.github.io/phiper/

Contributing

Bug reports, feature requests, and questions are welcome. Please open an issue on the GitHub tracker:

https://github.com/Polymerase3/phiper/issues