For clustering peptides based on shared residues and physicochemical properties, I recommend using GibbsCluster 2.0. This tool is specifically designed for unsupervised clustering and alignment of peptide sequences. It allows you to control the number of clusters and set parameters for identifying motifs within the peptide sequences. GibbsCluster can handle data of varying lengths and provides options to fine-tune clustering by focusing on subtle sequence differences or grouping peptides with similar properties.

Additionally, IEDB’s Epitope Cluster Analysis offers peptide clustering based on sequence identity. You can specify sequence identity thresholds and customize clustering parameters to suit your needs. This tool ensures that peptides sharing significant similarities are grouped together, providing insights into their biological relevance.

Both tools are highly customizable, and they will help you avoid generating artificial clusters typically seen in tools like ClustalW, offering better biological insights for your peptides.

Thanks so much Azmain! I have tried both, so I think I am on the right path. I liked the IEDB's better, but we will also try Hammonck (Bioinformatics, Volume 32, Issue 1, January 2016, Pages 9–16, https://doi.org/10.1093/bioinformatics/btv522) that seems nice. Kind regards, Emmanuel