Welcome to the LIgO documentation!

LIgO is a Python tool for simulation of adaptive immune receptors (AIRs) and repertoires (AIRRs) with known ground-truth immune signals for the development and benchmarking of AIRR-based machine learning. To get started using LigO right now, check out our Quickstart tutorial.

_images/ligo_pipeline.png

Why should you use LIgO?

  • LIgO makes simulations reproducible, all simulation parameters are specified through a YAML file.

  • LIgO contains different types of immune events and immune signals, including (gapped) k-mers, PWMs, specific V and J genes.

  • LIgO supports simulation of signal-specific AIRs using rejection sampling or signal implantation and preserves the AIRR generation probability distribution.

  • LIgO simulates synthetic AIRR (BCRs and TCRs) data both on the receptor and repertoire level, the single and paired-chain level.

  • LIgO guides the user and helps to set optimal simulation parameters.

  • LIgO outputs detailed information about presence and position(s) of immune signal(s) for every AIR in AIRR-compliant format.

Please check out LIgO manuscript (biorxiv link) for more information!

Contents