LSalign interpretations with large RMSDs

[lam.a]

Hello,

I was wondering if anyone has any experience with using LS-align on similar chemical analogues but had large RMSD values? I have two very similar compounds used on LS-align (differing only by the addition of a Fluorine atom) but the RMSD was ~3 for both the rigid and flexible LS-align options, shouldn't I be expecting an RMSD value closer to 0?

I retrieved these .mol2 files based on SMILES using https://datascience.unm.edu/tomcat/biocomp/convert, with the +2D and +3D boxes checked. I have also tried checking only the +2D option and results were better but still not expected (RMSD ~1)

These were the two SMILES I used:
Fc1ccc(cc1)c1nc([nH]c1c1ccncc1)c1ccc(cc1)S(=O)C
FC1=CC=C(C=C1)C(N=C2C3=CC=C(C(O)=O)C=C3)=C(N2)C4=CC=NC=C4