CAGEcleaner reduces redundancy in gene cluster mining hit sets. The redundancy in typical genome mining target databases (e.g. NCBI nr) often propagates into the result set, requiring extensive manual curation to carry out downstream analyses and visualisation efficiently.
Starting from a session file from a cblaster, CAGECAT, or ✨ cfoldseeker ✨ (our new protein structure similarity-based tool), CAGEcleaner dereplicates the hits based on a representative sample of the sequence regions that encode these hits (either full genomes or only the direct genomic neighbourhood). In addition, CAGEcleaner can automatically retain additional hits associated with non-representative sequences if they exhibit significant diversity in gene cluster contents or sequence similarity. Finally, CAGEcleaner returns a filtered cblaster session file and hit table, ready for downstream analyses.
Tip
Although CAGEcleaner can be used as a stand-alone tool, it is the dereplication engine of the ✨ csuite ✨, our new integrated toolbox featuring streamlined workflows for both sequence and protein structure-based gene cluster mining. Try it out!
- Full genome hit dereplication: Dereplicates the full genome assemblies of the host organisms using an ANI-based approach via
skDER, and retains the hits that are encoded by a representative genome assembly. The more conservative option that also takes the diversity of the host organism into account. Choose this option if you're concerned about preserving host diversity during compression, for example to identify HGT events. - Neighbourhood hit dereplication: Extracts a genomic region of a predefined length around each hit, clusters all extracted regions by sequence similarity using
MMseqs2, and retains the hits associated with the representative genomic regions. The more aggressive option that ignores host diversity. Choose this option if losing host diversity is not an issue. - Non-cblaster input:
CAGEcleanerhas originally been designed to use together withcblaster, but now it supports output from other mining tools too by supplying your hits as multiple formatted TSV files. See the docs and the example output for the specific formatting.
For installation instructions, usage, explanations and more, head over to the CAGEcleaner docs!
Note
CAGEcleaner only supports the single-mode cblaster modes (remote, local, hmm). We do not recommend using sessions from one of the combi modes.
Important
CAGEcleaner has no direct Windows support. If you have a seemingly successful installation directly on your Windows system, you likely have installed v1.1.0, an old version with known bugs! There are alternative options to run CAGEcleaner on Windows.
If you found CAGEcleaner useful, please cite our manuscript:
De Vrieze, L., Biltjes, M., Lukashevich, S., Tsurumi, K., Masschelein, J. (2025) CAGEcleaner: reducing genomic redundancy in gene cluster mining. Bioinformatics https://doi.org/10.1093/bioinformatics/btaf373
CAGEcleaner relies heavily on the following tools, so please give these proper credit as well.
Salamzade, R., & Kalan, L. R. (2025). skDER and CiDDER: two scalable approaches for microbial genome dereplication. Microbial Genomics, 11(7), https://doi.org/10.1099/mgen.0.001438
Shaw, J., & Yu, Y. W. (2023). Fast and robust metagenomic sequence comparison through sparse chaining with skani. Nature Methods, 20(11), 1661–1665. https://doi.org/10.1038/s41592-023-02018-3
Steinegger, M., & Söding, J. (2017). MMseqs2 enables sensitive protein sequence searching for the analysis of massive data sets. Nature Biotechnology, 35, https://doi.org/10.1038/nbt.3988
CAGEcleaner is freely available under an MIT license.
Use of the third-party software, libraries or code referred to in the References section above may be governed by separate terms and conditions or license provisions. Your use of the third-party software, libraries or code is subject to any such terms and you should check that you can comply with any applicable restrictions or terms and conditions before use.