Missmatch between InChI from SDF file export and InChI computetd using OpenBabel

[drjahu]

Dear LotusDB team!
I am sorry if this question is trivial and has an obvious answer!

Using the SDF file export 'LOTUS_2021_03_simple.sdf' I noticed that there are missmatches between the given InChI in the file and an InChI computet from the stucture (I use OpenBabel for the computatiion).

Please consider as an example the structure LTS0000001
InChI read from File and given on the website (https://lotus.naturalproducts.net/compound/lotus_id/LTS0000001):
InChI=1S/C28H46N4O6/c1-19(2)12-10-8-6-5-7-9-11-13-25(36)32-26(21(4)33)28(38)31-23-18-22(34)16-17-29-24(35)15-14-20(3)30-27(23)37/h7,9,11,13-15,19-23,26,33-34H,5-6,8,10,12,16-18H2,1-4H3,(H,29,35)(H,30,37)(H,31,38)(H,32,36)/b9-7+,13-11-,15-14-/t20-,21+,22+,23-,26-/m0/s1

InChI computet using OpenBabel:
InChI=1S/C28H46N4O6/c1-19(2)12-10-8-6-5-7-9-11-13-25(36)32-26(21(4)33)28(38)31-23-18-22(34)16-17-29-24(35)15-14-20(3)30-27(23)37/h7,9,11,13-15,19-23,26,33-34H,5-6,8,10,12,16-18H2,1-4H3,(H,29,35)(H,30,37)(H,31,38)(H,32,36)/t20-,21+,22+,23-,26-/m0/s1

The InChI read from file does not include the "/b9-7+,13-11-,15-14-" describing the /b sub-layer encoding the double bond stereochemistry.

As I am interested in both, the structure from the SDF file and an most deterministic InChI, I seam not to be able to get both information.

Is there any possibility to get an original SDF exports from LotusDB where the InChI computet from the compound structure encoded in SDF file matches the given InChI code?

I highly appreciate your time and effort.
Looking forward to a short answer!

Best regards,
Jan