Converting atoms with high radical counts to InChI generates incorrect results

[greglandrum]

Configuration:

• RDKit Version: all, tested on 2020.03.5
• Operating system: all

Description:

Here's the demo:

In [20]: m = Chem.MolFromSmiles('[C]')                                                                                                              

In [21]: m.Debug()                                                                                                                                  
Atoms:
	0 6 C chg: 0  deg: 0 exp: 0 imp: 0 hyb: 4 arom?: 0 chi: 0 rad: 4
Bonds:

In [22]: Chem.MolToInchi(m)                                                                                                                         
[08:58:18] WARNING: Radical center type replaced: 5->3
Out[22]: 'InChI=1S/CH2/h1H2'

Working directly from a mol block is fine:

In [23]: print(Chem.MolToMolBlock(m))                                                                                                               

     RDKit          2D

  1  0  0  0  0  0  0  0  0  0999 V2000
    0.0000    0.0000    0.0000 C   0  0  0  0  0 15  0  0  0  0  0  0
M  END


In [24]: Chem.MolBlockToInchi(Chem.MolToMolBlock(m))                                                                                                
Out[24]: 'InChI=1S/C'

and we can parse the InChI correctly:

In [25]: nm = Chem.MolFromInchi(Chem.MolBlockToInchi(Chem.MolToMolBlock(m)))                                                                        

In [26]: nm.Debug()                                                                                                                                 
Atoms:
	0 6 C chg: 0  deg: 0 exp: 0 imp: 0 hyb: 4 arom?: 0 chi: 0 rad: 4
Bonds:

Thanks to @vfscalfani for noticing and reporting this.