NeighborFinder is an R package that implements a local network inference method designed for microbiome data. It enables the targeted discovery of direct neighbors around a species of interest.
This repository contains the files and code needed to reproduce the figures presented in the article: "NeighborFinder: an R package inferring local microbial network around a species of interest" (Sola, M. et al., 2025).
It also contains the intermediate files needed to evaluate NeighborFinder's performance, as well as the corresponding R script for calculating them from semi-synthetic datasets.

The files are organized according to the following tree structure:

│ compute_example_fig1.R │ compute_execution_time_supplementary.R │ compute_performance_supplementary.R ├───example │ ├───network_application_fig1.jpg │ └───res_all_8_datasets.rds ├───execution_time │ ├───exec_time_figS3.jpg │ └───time_high_prev_species_NF.rds ├───graphs ├───performances │ ├───res_n100 │ │ └───filtering_top_100 │ ├───res_n1000 │ │ └───filtering_top_100 │ ├───res_n250 │ │ └───filtering_top_100 │ ├───res_n50 │ │ └───higher_prevalence │ │ └───filtering_top_100 │ ├───res_n500 │ │ └───filtering_top_100 │ ├───scores │ │ ├───n100 │ │ ├───n1000 │ │ ├───n250 │ │ ├───n50 │ │ │ └───higher_prevalence │ │ └───n500 │ └───truth │ └───higher_prevalence └───performance_figures ├───heatmap_100-1000_figS1.jpg └─── heatmap_50_figS2.jpg

This repository has been organized with an appropriate tree structure to facilitate the execution of the R scripts. These scripts use intermediate files to reproduce the figures. Since generating them takes quite a long time, we have chosen to store all these intermediate files in zip folders.
In practice, users can download the entire repository and unzip the folders so that the paths are correct. To generate figures efficiently, users are invited to execute the part of the R scripts that depends on the files already prepared.