Skip to contents

bHIVE

B-cell Hybrid Immune Variant Engine

Overview

bHIVE is an R/Bioconductor package that implements a modular Artificial Immune System (AIS) for clustering and classification of high-dimensional data such as mass cytometry and single-cell profiles. Built on AI-Net (de Castro & Von Zuben 2001), bHIVE learns a set of prototypes (antibodies) by clonal selection, mutation and network suppression. The prototypes summarize cell populations in the original marker units, and the same fitted model labels new cells. Optional modules drawn from immunology add somatic hypermutation strategies, germinal center selection, microenvironment-driven adaptation and more.

Performance-critical operations (affinity/distance matrices, clonal selection, network suppression, mutation) are implemented in C++ via RcppArmadillo, with parallelization support through BiocParallel.

Key Features

  • Two tasks – clustering and classification on matrices, SummarizedExperiment and SingleCellExperiment objects
  • Bioconductor integration – labels stored in colData(), and bHIVEParam() runs bHIVE through bluster::clusterRows() and bluster::clusterSweep()
  • C++ backend – BLAS-optimized bulk affinity/distance computation
  • Two APIs – functional (bHIVE()) for quick use and R6 (AINet$new()) for full module composition
  • Multilayer architecture – honeycombHIVE() for hierarchical prototype refinement across layers
  • Hyperparameter tuning – swarmbHIVE() with grid search and BiocParallel
  • Gradient refinement – refineB() post-processing with 5 optimizers and several classification-aware loss functions
  • Composable immune modules – mix and match biological mechanisms via dependency injection
  • caret compatible – bHIVEmodel and honeycombHIVEmodel for cross-validation workflows

Installation

bHIVE is submitted to Bioconductor. Once accepted, install it with:

if (!requireNamespace("BiocManager", quietly = TRUE))
    install.packages("BiocManager")
BiocManager::install("bHIVE")

The development version is available from GitHub:

BiocManager::install("BorchLab/bHIVE")

Quick Start

Functional API

The simplest way to use bHIVE. Works like any R modeling function:

library(bHIVE)
data(iris)
X <- as.matrix(iris[, 1:4])

# Clustering
res <- bHIVE(X, task = "clustering", nAntibodies = 30, maxIter = 20)
table(res$assignments)

# Classification
res <- bHIVE(X, y = iris$Species, task = "classification",
             nAntibodies = 30, maxIter = 20)
table(Predicted = res$assignments, Actual = iris$Species)

Bioconductor objects

bHIVE() clusters the cells (columns) of a SingleCellExperiment using an assay or a reducedDim() embedding, and stores the labels in colData(). bHIVEParam() plugs bHIVE into bluster:

library(SingleCellExperiment)
sce <- SingleCellExperiment(assays = list(logcounts = t(X)))
reducedDim(sce, "PCA") <- prcomp(X)$x[, 1:3]

sce <- bHIVE(sce, use.dimred = "PCA", targetK = 3, verbose = FALSE)
table(sce$bHIVE, iris$Species)

colLabels(sce) <- bluster::clusterRows(
  reducedDim(sce, "PCA"),
  bHIVEParam(targetK = 3, affinityParams = list(alpha = "auto"))
)

See vignette("bHIVE") for a full analysis of mass cytometry data, including a comparison with FlowSOM and graph-based clustering.

R6 API with Modules

For full control, compose an AINet with any combination of immune modules. Each module is optional and slots into the fit loop at the appropriate biological stage:

# Adaptive mutation + germinal center selection
model <- AINet$new(
  nAntibodies = 20,
  maxIter = 30,
  shm = SHMEngine$new(method = "adaptive", base_rate = 0.1),
  germinalCenter = GerminalCenter$new(nTfh = 20, selectionPressure = 0.2),
  verbose = FALSE
)
model$fit(X, iris$Species, task = "classification")
table(model$result$assignments)

# Predict on new data
preds <- model$predict(X[1:10, ])

A richer composition uses microenvironment-aware exploration, Tfh-mediated quality selection, density-driven isotype switching, and a persistent memory pool:

me  <- Microenvironment$new()                              # zone classification
cs  <- ClassSwitcher$new(alpha_IgM = 0.1, alpha_IgG = 5)   # zone -> kernel width
gc  <- GerminalCenter$new(nTfh = 8, selectionPressure = 0.6)
mp  <- MemoryPool$new(archive_threshold = 0.05)            # persists across fits

model <- AINet$new(
  nAntibodies = 30, maxIter = 20,
  shm              = SHMEngine$new(method = "hotspot"),
  microenvironment = me,
  classSwitcher    = cs,            # requires microenvironment
  germinalCenter   = gc,
  memory           = mp,
  verbose          = FALSE
)
model$fit(X, iris$Species, task = "classification")

# Memory persists -- a second fit will recall relevant cells (clustering only)
# and continue archiving high-affinity antibodies.
mp$size()

Architecture

Algorithm

bHIVE evolves a population of antibody vectors to represent structure in data. Each iteration runs a subset of the following stages — modules attach to specific stages and are skipped when absent:

  1. Initialization (once) – sample from data, random generation, kmeans++, or V(D)J combinatorial assembly via VDJLibrary. If a MemoryPool carries cells from a prior fit, relevant memory is recalled and merged into the starting repertoire (clustering only).
  2. Activation gating – ActivationGate sets aside antibodies in over-dense neighborhoods or below an affinity floor so clonal selection runs on the sparse subset; gated antibodies rejoin after.
  3. Clonal selection + SHM – top-k antibodies cloned per data point. Mutation dispatches through SHMEngine (uniform, airs, hotspot, energy, adaptive); the adaptive strategy threads per-antibody Adam-style moment matrices across iterations.
  4. Germinal center selection – GerminalCenter runs Tfh-mediated quality selection; survivors weighted by clustering compactness or classification purity.
  5. Microenvironment & class switching – Microenvironment classifies each antibody into stable / explore / boundary zones and applies density-dependent jitter. ClassSwitcher binds each zone to an isotype (IgM/IgG/IgA) and sets the next iteration’s kernel width.
  6. Idiotypic regulation (experimental) – IdiotypicNetwork runs bell-shaped Ab-Ab dynamics that cull both over-clumped and isolated antibodies in small, tight repertoires. The rule acts on each antibody pair, so in large or spread-out repertoires it removes every antibody; AINet then keeps the top 10% by population and reports it. See ?IdiotypicNetwork.
  7. Network suppression – removes near-duplicate antibodies within an epsilon-ball under the chosen distance metric.
  8. Orphan pruning + final assignment (once) – antibodies that bind no training point are dropped; remaining cells produce cluster IDs or class predictions. MemoryPool archives high-affinity survivors back into the pool.

Immune Modules

Each module is an R6 class that can be injected into AINet via its constructor. All modules are optional – use only what you need.

Module Biological Basis What It Does
SHMEngine Somatic hypermutation 5 mutation strategies: uniform, airs, hotspot, energy, adaptive
IdiotypicNetwork Ab-Ab network regulation Bell-shaped activation dynamics replacing epsilon-threshold suppression (experimental)
GerminalCenter Tfh-B cell interaction Task-aware quality selection with resource competition
Microenvironment Tissue microenvironment cues Density-dependent zone classification and mutation rate modulation
VDJLibrary V(D)J recombination Combinatorial gene library initialization (PCA, cluster, random partition)
ActivationGate Two-signal activation Costimulatory filtering (density, danger signal, or label entropy)
MemoryPool Immunological memory Archive high-affinity antibodies and recall on distribution shift
ClassSwitcher Isotype class switching IgM (broad) / IgG (specific) / IgA (boundary) kernel width modulation
ConvergentSelector Public clonotypes Cross-repertoire consensus for ensemble methods

Multilayer & Tuning 

# honeycombHIVE: hierarchical prototype refinement
res <- honeycombHIVE(X, y = iris$Species, task = "classification",
                     layers = 3, nAntibodies = 30,
                     refine = TRUE, refineOptimizer = "adam")

# swarmbHIVE: hyperparameter grid search (parallelizable)
grid <- expand.grid(nAntibodies = c(15, 30), beta = c(3, 5), epsilon = c(0.01, 0.1))
best <- swarmbHIVE(X, y = iris$Species, task = "classification",
                   grid = grid, metric = "accuracy", maxIter = 20)
best$best_params

Gradient Refinement

Fine-tune antibody positions after training with refineB():

res <- bHIVE(X, y = iris$Species, task = "classification",
             nAntibodies = 20, maxIter = 20)

# Adam-based refinement with cross-entropy loss. antibody_index gives the
# antibody (row of res$antibodies) that labeled each sample.
A_refined <- refineB(res$antibodies, X, y = iris$Species,
                     assignments = res$antibody_index,
                     task = "classification",
                     loss = "categorical_crossentropy",
                     optimizer = "adam", steps = 10, lr = 0.01)

Affinity & Distance Functions

Affinity Formula Use Case
gaussian exp(-alpha ||x - a||^2) General purpose (default)
laplace exp(-alpha ||x - a||) Heavier tails than Gaussian
polynomial (x . a + c)^p Non-Euclidean similarity
cosine (x . a) / (||x|| ||a||) Direction-based similarity
hamming 1 - (mismatches / d) Categorical/binary features
Distance Notes
euclidean Default; L2 norm
manhattan L1 norm
minkowski Generalized Lp (parameter p)
cosine 1 - cosine similarity
mahalanobis Accounts for feature covariance (requires Sigma)
hamming Count of differing features

Bug Reports / Feature Requests

If you run into any issues or bugs please submit a GitHub issue with details of the issue. If possible please include a reproducible example. Any requests for new features or enhancements can also be submitted as GitHub issues.

Contributing

We welcome contributions to the bHIVE project! To contribute:

  • Fork the repository.
  • Create a feature branch (git checkout -b feature-branch).
  • Commit your changes (git commit -m "Add new feature").
  • Push to the branch (git push origin feature-branch).
  • Open a pull request.