Back

Meta-PepView: a metaproteomics performance evaluation and visualization platform

van der Zwaan, R.; van Olst, B.; van Loosdrecht, M. C.; Pabst, M.

2025-12-07 systems biology
10.64898/2025.12.03.692192 bioRxiv
Show abstract

Microbial community proteomics is rapidly gaining traction as it allows exploration of functional processes in microbial ecosystems. Consequently, there is a growing need for user-friendly tools that enable performance evaluation and interactive visualization of the increasingly complex community proteomics data. We introduce meta-PepView, a web-based platform that enables performance evaluation and interactive visualization of metaproteomics data, ensuring transparent and reproducible metaproteomics experiments. Meta-PepView integrates spectral sequencing outputs and databases from common proteomics search engines and classification tools. It runs efficiently on laptop or desktop PCs and can be deployed as a Docker container or installed with pip, from which it is operated through the web browser. Its modular design allows easy expansion with new data sources and annotation databases. Meta-PepView is an open-source Python platform that is freely available under the Apache License 2.0. The code is available on GitHub: https://github.com/ramonzwaan/metapepview/. Here, we showcase the meta-PepView platform with data from synthetic communities and previously published microbiome studies.

Matching journals

The top 2 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.