B1 parameter is incorrect for some substituents

[ajs99778]

To clarify, I used the command line Sterimol, but the algorithm it uses looks the same as wSterimol.

I was getting values for CHF₂ and CHCl₂ that seemed too small, so I added some code to Sterimol so it would plot a 2-D projection of the radii and some of the vectors Sterimol was trying to assign as B₁. Below is the projection of the radii for CHCl₂. Sterimol reports the B₁ as 1.64 for this structure (indicated by the red dot), but this is not a valid B₁. The FORTRAN sterimol program gets 1.88, so Sterimol should probably be choosing the orange or purple dots.

Here's the structure I used in .com format:

#n AM1 opt 

steri13.xyz

0 1
H      0.00000     0.00000     0.00000
C     -1.10000     0.00000     0.00000
Cl    -1.69000     1.17000     1.17000
Cl    -1.69000     0.43000    -1.60000
H     -1.47000    -1.00000     0.27000

This structure is not optimized, but optimizing doesn't change the results very much.

command used:

python -m sterimol -a1 1 -a2 2 CHCl2.com

relevant output:

   STERIMOL: using cpk van der Waals parameters

    Structure                      L        B1        B5
   CHCl2.com                   3.89      1.64      3.46

[luchini18]

Hi Tony,
Thanks for noticing this - these code packages (wSterimol and Sterimol) are currently not being actively developed, however, all new developments to Python-based Sterimol parameters from our group now take place in DBSTEP and it looks as though this is no longer an issue. The new package will allow you to collect these Sterimol parameters on the command line or through a Python script, along with a few other steric parameters, also allowing for visualization of the parameters measured using the PyMOL program.

I loaded in your example CHCl2 molecule for comparison. Using DBSTEP you can measure steric parameters using van der Waals radii (as opposed to CPK radii used in the Sterimol code) or using a grid-based method. You are able to switch between these with "--measure classic" (for VDW radii) or "--measure grid" (grid is the default).

Using this command measure the Sterimol parameters from VDW radii
python -m dbstep chcl2.xyz --atom1 1 --atom2 2 --sterimol --measure classic
I receive the output
chcl2.xyz / Bmin: 1.93 / Bmax: 3.41 / L: 3.44
(The slight value difference from your results is attributed to a difference in CPK vs VDW radii)

I've attached the output visualization of the result below, and it looks like the Bmin parameter (in green) is corresponding to a value near where you had suggested the correct value should be.

The Sterimol repository has been archived for a couple of years, but we will update wSterimol to use DBSTEP, which should resolve the underlying issue for this package.