mBatchNet: an interactive web server for diagnosis, correction, and benchmarking of batch effects in microbiome data

Batch-effect diagnosis and correction are important for reproducible microbiome analysis and cross-study integration. Several batch-correction algorithms are available, but applying and comparing established methods in practice remains nontrivial because they differ in assumptions, accepted inputs, parameters, and evaluation outputs. Here, we present mBatchNet, an interactive web server for applying established batch-correction methods to processed microbiome feature tables and evaluating their effects within a single workflow. The server supports correction methods spanning recent microbiome-oriented approaches and established general-purpose baselines, validates uploaded feature tables and metadata, flags batch-target association, applies matched pre- and post-correction diagnostics, and exports corrected matrices, statistical summaries, run logs, and reproducibility records. In a 16S ribosomal RNA (rRNA) anaerobic digestion case study, mBatchNet revealed method-dependent differences in batch attenuation and phenotype preservation, highlighting its utility for comparing correction strategies.

mBatchNet is freely available without login at https://mbatchnet.com/. The latest source code is available at https://github.com/gilmore307/mBatchNet, and is archived at https://doi.org/10.5281/zenodo.20767444. The server is implemented with a Python/Dash front end and coordinated Python/R back-end analysis scripts.