BCUT mismatch??

User 21b7e0228c

10-03-2009 09:56:42

Hello there!

I was routinely using BCUT descriptor generation in my QSAR models, though I have to admit that I never dug into the details. I was just using generatemd 3.2.3 to spit out default BCUTSs - it used to return 4 columns, but I never cared which BCUTs those were, as far as it did its job reproducibly (I would have cared if these would have been entering the final QSAR models, but that did not happen).

However, I now notices that, upon upgrading to generatemd 5.0.0 only ONE BCUT column is generated. How can I recreate a config file letting generatemd 5.0.0 behave again like its predecessor (I still don't wanna know the gory details - I'm in an agnostic phase ;-)



ChemAxon efa1591b5a

10-03-2009 16:00:47

Hi Dragos,

that's rather odd. It should produce 4 values, the 2 lowest and the 2 highest eigenvalues...

There was a bug in BCUT generation, it was fixed in version 5.0.6. Anyway, that should not affect the number of values produced. I tried v 5.0 as well as the current version to be released very soon and both worked well.

Could you tell us the exact version number of your JChem installation and copy your full generatemd command line in your post?

Thank you and apologies for any inconvenience this problem caused.

Best regards,


User 21b7e0228c

10-03-2009 16:15:40

Well, as far as I can tell it's all 5.0.0. Command options were as dumb as possible

c -k BCUT -D

If this doesn't help, I can let you log on on my machine. It is strange, indeed!

ChemAxon efa1591b5a

11-03-2009 09:09:44

That was another issue fixed in 5.0.6. I checked 5.0.0 release version and it outputs one value only, as you experienced. (Previously I checked my local copy of 5.0 which apparently was not the same as 5.0.0 release - apologies for this mistake.)

So, the conclusion is that you need to upgrade to at least 5.0.6, I'm afraid. That should solve the problem.

Best regards,


User 21b7e0228c

11-03-2009 11:12:39

Thank you!

ChemAxon efa1591b5a

11-03-2009 11:15:22

Great, thank you.