Hi everyone! I’m currently analyzing leucine-rich repeats (LRRs) across a set of related proteins and need a reliable way to extract individual LRR units. I’m torn between two approaches:

1. Using HMMER (hmmscan) with a single LRR HMM and then piecing together consecutive hits from the output. My concern here is how to accurately merge adjacent or partial hits to form full, contiguous repeats.

2. Using HHrepID (https://doi.org/10.1093/bioinformatics/btn039) to detect tandem repeats in each protein. This seems straightforward for identifying repeats, but I’m worried about inconsistent boundaries across different proteins—making alignment and cross-comparison more difficult.

Because these are phaeophyceae proteins, some standard LRR prediction tools don’t seem to perform that well. Has anyone else encountered this challenge? Do you have tips or alternative tools that handle boundary detection more consistently for highly variable LRRs? Thank you in advance!