A couple of followups here since I had some time on a plane yesterday.
On Wed, Jul 15, 2015 at 11:02 AM, Greg Landrum <[email protected]>
wrote:
>
> On Tuesday, July 14, 2015, Michael Reutlinger <[email protected]> wrote:
>
>>
>> thanks for your answer. I also figured that the invariant feature of the
>> AtomPair functions should be well suited. Initially, I thought that the
>> pi-electron and neighbouring atom count would always be added, but this is
>> clearly not the case.
>>
>
> It shouldn't be the case for the AP fingerprint. It *does*, however, look
> like there will be a problem with the TT fingerprints if you provide your
> own invariants (this is based on a skim of the code, so I'm not sure that's
> correct).
>
I skimmed too quickly before writing this response. User-provided atom
invariants should also work properly with TT descriptors. The branching
terms added in the TT calculation do seem to be properly handed with
user-provided invariants.
> Btw. is the hashing method used for creating the hashed fingerprint still
>> valid / suitable if only a small number of the maximal
>> numAtomPairFingerprintBits
>> / codeSize is actually used?
>>
>
> The quick read-through I did of the code makes me believe that it should
> be fine. I will confirm over the next day or so and let you know.
>
Looks like it should be fine.
-greg
------------------------------------------------------------------------------
Don't Limit Your Business. Reach for the Cloud.
GigeNET's Cloud Solutions provide you with the tools and support that
you need to offload your IT needs and focus on growing your business.
Configured For All Businesses. Start Your Cloud Today.
https://www.gigenetcloud.com/
_______________________________________________
Rdkit-discuss mailing list
[email protected]
https://lists.sourceforge.net/lists/listinfo/rdkit-discuss
