Back

ABC_GP_IPM:A Python package for applying GP and ABC to Integral Projection Models, with a Soay sheep case study

Zhu, Z.; Christodoulou, M. D.; Steinsaltz, D.

2026-01-09 evolutionary biology
10.64898/2026.01.08.698488 bioRxiv
Show abstract

The integration of Gaussian Process (GP) models with Approximate Bayesian Com-putation (ABC) has been explored as a flexible framework for constructing Integral Projection Models (IPMs), enabling non-parametric modelling of demographic rela-tionships and the incorporation of population-level information without explicit like-lihoods. However, the practical implementation of this framework - particularly the selection of ABC summary statistics and the execution of ABC-PMC samplers - re-mains non-trivial and can limit its broader adoption. To address this gap, we introduce ABC_GP_IPM, a Python package that provides a streamlined and user-friendly interface for constructing GP- and ABC-based IPMs. We demonstrate the utility and performance of the package through a real-world case study of Soay sheep (Ovis aries), illustrating how the software simplifies complex modelling workflows and data management in IPMs.

Matching journals

The top 7 journals account for 50% of the predicted probability mass.

50% of probability mass above

"Similar papers" are the closest papers from that journal in the model's embedding space. They show what the match is built on, but the ranking comes mostly from a classifier over the whole training set, not from these examples alone.