RMDetect fails to find modules in a given sequence

[tzok]

RMDetect fails to recognize any motif in the example lysine riboswitch sequence:

$ python2 rmdetect.py --data-path models --out lys_result.txt examples/lysine/lys.seq
Searching for models in:
	1. '--data-path' option: OK! -> 'models'

Loading models from: 'models' ....
Included models:  TGA_1.0, KT_1.0, GB_1.0, CL_1.0
Excluded models:  none
Scanning 1 sequences...
Sequence 1 of 1 'unique'slide pos 76 of 194
DONE
$ python2 rmout.py --data-path models < lys_result.txt
Searching for models in:
	1. '--data-path' option: OK! -> 'models'

Loading models from: 'models' ....
Included models:  TGA_1.0, KT_1.0, GB_1.0, CL_1.0
Excluded models:  none
Reading data from results file: None

- - - - - - - - - - - - - - - - - - - -
Found 0 candidates:

    # sequence model ver. score bpp pos cols motif
- - - - - - - - - - - - - - - - - - - -

It seems that RMDetect correctly loads models for kink-turn and G-bulged modules, but is unable to find them in the given sequence. This is incorrect according to supplementary materials of the main RMDetect paper: https://static-content.springer.com/esm/art%3A10.1038%2Fnmeth.1603/MediaObjects/41592_2011_BFnmeth1603_MOESM173_ESM.pdf

According to it, the output should be:

- - - - - - - - - - - - - - - - - - - -
Found 2 candidates:
 # sequence model ver. score bpp pos cols motif
 1 unique KT 1.0 14.27 0.45 35-51 35-51 GUGAGG-CCAGAGA.CC
 2 unique GB 1.0 6.88 0.09 23-65 23-65 CAGGUA.G-UGAGGG
- - - - - - - - - - - - - - - - - - - - 

What is the reason? Why cannot RMDetect find the modules?