How to quantify immigration from community abundance data using the neutral community model

Area of Science:

• Ecology
• Computational Biology
• Bioinformatics

Background:

• Biological communities are interconnected via dispersal, influencing local and regional diversity.
• Direct measurement of dispersal is challenging, hindering understanding of its ecological impact.
• The Neutral Community Model (NCM) offers a way to estimate immigration rates (N_Tm) using abundance data.

Purpose of the Study:

• To evaluate the accuracy of NCM-based immigration rate (N_Tm) estimation.
• To introduce and compare novel inference methods for N_Tm.
• To determine optimal sampling strategies for accurate N_Tm estimation in diverse communities.

Main Methods:

• Introduced two new inference methods: variance-based and Dirichlet-multinomial log-likelihood (DM-LL).
• Complemented these with the established occupancy-based inference method.
• Validated methods using simulations of activated sludge microbiomes and real-world datasets (wastewater, tropical trees, coral reefs).

Main Results:

• All tested methods estimated N_Tm within 10% of ground-truth values in simulations.
• Variance-based and DM-LL methods required less sampling effort for accurate estimates.
• Accurate N_Tm inference necessitates read depths exceeding the immigration rate per sample.

Conclusions:

• Developed and validated robust methods for quantifying community immigration using NCM.
• Identified key sampling requirements (read depth) for reliable N_Tm estimation.
• Provided practical guidelines for studying dispersal in complex, diverse biological communities.