immunomind/immunarch

🧬 immunarch [R package] – Multi-Modal Immune Repertoire Analytics for Immunotherapy and Vaccine Design

What it solves

immunarch is a toolkit for analyzing Adaptive Immune Receptor Repertoire (AIRR) data. It addresses the challenge of processing massive, multi-modal immune profiling data (such as single-cell V(D)J, spatial transcriptomics, and clinical metadata) to extract meaningful biological insights and biomarkers for personalized immunotherapies and vaccines.

How it works

The package provides a modular R-based pipeline for computing receptor- and repertoire-level statistics. It leverages immundata to handle datasets that exceed system memory, allowing the same code to run on local machines or servers. The toolkit includes tools for diversity analysis, clonality analysis, and public receptor indices, and it provides an extension API and adapters to interoperate with other AIRR tools.

Who it’s for

It is designed for bioinformatics researchers and clinicians who analyze T-cell and B-cell immune repertoires to discover biomarkers and design personalized medical treatments.

Highlights

  • Multi-modal profiling: Integrates single-cell, spatial, and immunogenicity annotations.
  • Scale: Handles datasets from gigabytes to terabytes without requiring code changes for server migration.
  • ML-Ready: Builds feature tables for downstream machine learning and deep learning models.
  • Comprehensive Statistics: Built-in functions for gene usage, diversity (Pielou, Shannon, Chao1), and clonality (prop).

Related

  • Project
    enoche/MMRec

    A modern multimodal recommendation toolbox that provides a unified framework and source code for a wide variety of research-based recommendation models.

  • Project
    pyRiemann/pyRiemann

    A machine learning package for processing and classifying multivariate data using the Riemannian geometry of symmetric positive definite matrices, primarily for biosignals and remote sensing.

  • Project
    MarioniLab/miloR

    miloR is an R package for differential abundance analysis on single-cell datasets using k-nearest neighbor graphs to identify changing cell populations.

  • Project
    PreferredAI/cornac

    A comparative framework for multimodal recommender systems that simplifies the integration of auxiliary data and the benchmarking of various recommendation algorithms.

  • Project
    sunlabuiuc/PyHealth

    A comprehensive deep learning toolkit for clinical predictive modeling that provides a modular pipeline for healthcare AI applications.