Re: [Rdkit-discuss] How to decompose the UFF (or MMFF94) scoring of a small molecule?

[Geoffrey Hutchison]

But it’s not so simple. If you rotate a dihedral, all the non-bonded VdW 
interactions also change.

Calculating a force field energy at one particular geometry is extremely fast, 
and the current version allows you to run multiple energies in parallel.

Generate a bunch of geometries at once and then score those in parallel. Trying 
to “update” isn’t worth it, because there are always more non-bonded 
interactions (n^2) than bonds and angles.

-Geoff
On May 18, 2023 at 4:24 AM -0400, Francois Berenger , wrote:
> Dear list,
>
> I asked this question in rdkit's github discussions:
>
> https://github.com/rdkit/rdkit/discussions/6377
>
> But, apparently that's not more responsive than the ML, so here is my
> question:
> ---
> I have a ligand, I would like to score its current conformer using
> rdkit's UFF implementation.
>
> Later, I would like to change some rotatable bond (single bond out of
> ring) and update
> the conformer's energy bu just re-evaluating the part of the energy that
> is supposed to have
> changed (i.e. the dihedral component).
> Bond lengths, bond angles and partial charges being constant.
>
> I suspect it should be faster than rescoring the conformer from scratch.
>
> How to do this with rdkit?
> ---
>
> Thanks a lot,
> Francois.
>
>
> ___
> Rdkit-discuss mailing list
> Rdkit-discuss@lists.sourceforge.net
> https://lists.sourceforge.net/lists/listinfo/rdkit-discuss
___
Rdkit-discuss mailing list
Rdkit-discuss@lists.sourceforge.net
https://lists.sourceforge.net/lists/listinfo/rdkit-discuss

Re: [Rdkit-discuss] UFF and MMFF94 strongly disagree on the internal energy of one conformer

[Geoffrey Hutchison]

>
> I am a little bit annoyed by the fact that the UFF energy
> is not even negative and that the two FFs disagree by so much.
>

I would not consider this to be a big problem, IMHO.

Do the two FFs compute something different?
>

Yes, very much so. UFF, for example does not include electrostatic
non-bonded terms, and MMFF94 does. Additionally, MMFF94 includes a specific
term for hydrogen bonding interactions.

I consider force field energies to only be relevant in relative terms. That
is, taking the relative difference of energy of relative conformers using a
force field is more useful than the absolute value. (And particularly
comparing between numbers from UFF vs. MMFF94 vs. GAFF, etc.)

Even then, while force fields are useful for “quick cleanup” as my group
showed (https://doi.org/10.1002/qua.26381) the relative energy rankings of
UFF and MMFF94 are still crude relative to ML or quantum methods.

In other words, the question is “does optimizing a conformer with UFF or
MMFF94 help?”

Yes. In general, conformers generated using RDKit (or Open Babel .. or most
methods I’ve seen) have lower RMSD when optimized by a force field.

Hope that helps,
-Geoff

---
Prof. Geoffrey Hutchison
Department of Chemistry
University of Pittsburgh
tel: (412) 648-0492
email: geo...@pitt.edu
twitter: @ghutchis
web: https://hutchisonlab.org/
___
Rdkit-discuss mailing list
Rdkit-discuss@lists.sourceforge.net
https://lists.sourceforge.net/lists/listinfo/rdkit-discuss

Re: [Rdkit-discuss] UFF rdkit implementation: QEq partial charges are missing?

[Geoffrey Hutchison]

Dear Francois,

Actually, the UFF implementation doesn’t include partial charges. The paper 
says something like “you could use QEq charges” but in fact every UFF 
implementation I’ve seen, including from the Goddard group doesn’t include them.

My understanding is that the parameters were derived without charges to make it 
easier to handle periodic boundary conditions (ie no need for Ewald sums and 
the like).

So Open Babel can compute QEq charges, but they’re not included in a UFF 
calculation.

I actually use this as a demo in Avogadro… put a few water molecules and UFF 
doesn’t do much. Switch to MMFF94 and suddenly they reorient and  hydrogen bond.

I’ve always wanted to see a modern parameterization of UFF with polarizable 
electrostatics, etc.

Best,
Geoff
On Mar 16, 2023 at 4:13 AM -0400, Francois Berenger , wrote:
> Dear list,
>
> Am I right that rdkit has an UFF implementation, but the
> recommended partial charges for UFF cannot be computed by rdkit?
> Those charges are called QEq in the UFF seminal paper[1].
>
> I think openbabel can compute them, but I am surprised if this is not
> included in rdkit.
>
> Does it mean that the UFF rdkit implementation doesn't include
> electrostatic
> interactions?
>
> Will hardcore computational chemists send me to hell if I use UFF in
> combination
> with the MMFF94 partial charges?
>
> Should we include QEq charges in rdkit? (I honestly don't know yet if
> this is hard
> to implement)
>
> Thanks a lot,
> Francois.
>
> [1] UFF, a full periodic table force field for molecular mechanics and
> molecular dynamics simulations
> https://pubs.acs.org/doi/abs/10.1021/ja00051a040
>
>
> ___
> Rdkit-discuss mailing list
> Rdkit-discuss@lists.sourceforge.net
> https://lists.sourceforge.net/lists/listinfo/rdkit-discuss
___
Rdkit-discuss mailing list
Rdkit-discuss@lists.sourceforge.net
https://lists.sourceforge.net/lists/listinfo/rdkit-discuss

Re: [Rdkit-discuss] new RDKit FAQ

[Geoffrey Hutchison]

>
> - Why (and when) do I need to add Hydrogens to molecules
> - How to perfom 3D generation from SMILES (or other)
>

For example “Do I really need to add hydrogens before 3D generation?”

Cheers,
-Geoff
___
Rdkit-discuss mailing list
Rdkit-discuss@lists.sourceforge.net
https://lists.sourceforge.net/lists/listinfo/rdkit-discuss

Re: [Rdkit-discuss] Using DrawAttachmentLine for bidentate ligands

[Geoffrey Hutchison]

 Yep, I’m aware .. but it wouldn’t take too much work to at least ignore
“<“ and “>” in the Open Babel parser. 

In the meantime, my Python scripts will strip them from the SMILES strings
before passing them to Open Babel.

-Geoff

On Aug 14, 2022 at 4:42:25 AM, David Cosgrove 
wrote:

> Hi Geoff,
> You should bear in mind that the dative bond syntax is an RDKIt extension
> to SMILES so is not guaranteed to be parsed correctly by other
> cheminformatics toolkits.
> Dave
>
>
> On Sun, 14 Aug 2022 at 00:42, Geoffrey Hutchison <
> geoff.hutchi...@gmail.com> wrote:
>
>> mol = Chem.MolFromSmiles("C12=CC=CC=[N]1->[*]<-[N]3=C2C=CC=C3")
>>
>>
>> Hmm. I forgot the SMILES syntax of dative bonds. That's a nice idea.
>>
>> I actually decided to use the noFreeType=True option, add a highlight as
>> the "metal" and remove the * from the depiction.
>>
>> The initial set is at:
>> https://github.com/OpenChemistry/fragments/tree/main/ligands
>>
>> Thanks!
>> -Geoff
>>
>>
>>
>>
>> ___
>> Rdkit-discuss mailing list
>> Rdkit-discuss@lists.sourceforge.net
>> https://lists.sourceforge.net/lists/listinfo/rdkit-discuss
>>
> --
> David Cosgrove
> Freelance computational chemistry and chemoinformatics developer
> http://cozchemix.co.uk
>
>
___
Rdkit-discuss mailing list
Rdkit-discuss@lists.sourceforge.net
https://lists.sourceforge.net/lists/listinfo/rdkit-discuss

Re: [Rdkit-discuss] Using DrawAttachmentLine for bidentate ligands

[Geoffrey Hutchison]

> mol = Chem.MolFromSmiles("C12=CC=CC=[N]1->[*]<-[N]3=C2C=CC=C3")

Hmm. I forgot the SMILES syntax of dative bonds. That's a nice idea.

I actually decided to use the noFreeType=True option, add a highlight as the 
"metal" and remove the * from the depiction.

The initial set is at: 
https://github.com/OpenChemistry/fragments/tree/main/ligands 


Thanks!
-Geoff




___
Rdkit-discuss mailing list
Rdkit-discuss@lists.sourceforge.net
https://lists.sourceforge.net/lists/listinfo/rdkit-discuss

[Rdkit-discuss] Using DrawAttachmentLine for bidentate ligands

[Geoffrey Hutchison]

I've been using RDKit for depicting sets of ligands from SMILES, which has been 
great.

I'd like to add some bidentate and tridentate ligands. Let's stick to 
bipyridine at first (see image)



I have the appropriate SMILES, leaving * as part of a 5 atom ring involving the 
nitrogen atoms (and sanitize = False)
C12=CC=CC=[N]1[*][N]3=C2C=CC=C3

The resulting depiction is great .. except I'd like to add the attachment 
"squiggles" across the N-* bonds.

I see there are DrawAttachmentLine and DrawWavyLine methods, but I'd need to 
get the positions of the N and * atoms.

What's the best way to do that and / or automate adding wavy lines across 
arbitrary bonds?

Thanks,
-Geoff

___
Rdkit-discuss mailing list
Rdkit-discuss@lists.sourceforge.net
https://lists.sourceforge.net/lists/listinfo/rdkit-discuss

Re: [Rdkit-discuss] SMARTS pattern

[Geoffrey Hutchison]

> Nevermind, x3 won't exclude the fused 4-atom rings from your first example. 
> I'll let you know if I think of some other way. :-)

I think you'd want something like this, perhaps - to exclude atoms in ring size 
4?

[*;R2!r4]~1~[*;R2!r4]~[*;R2!r4]~[*;R2!r4]~[*;R2!r4]~[*;R2!r4]~1

I also don't know if you're trying to ensure that each of the atoms are 
aromatic, in which case, you'd want something like:

[a;R2!r4]~1~[a;R2!r4]~[a;R2!r4]~[a;R2!r4]~[a;R2!r4]~[a;R2!r4]~1

Hope that helps,
-Geoff___
Rdkit-discuss mailing list
Rdkit-discuss@lists.sourceforge.net
https://lists.sourceforge.net/lists/listinfo/rdkit-discuss

Re: [Rdkit-discuss] SVG depiction with fonts?

[Geoffrey Hutchison]

> Unfortunately there are not keyword arguments for this (something for me to 
> fix ASAP), but you can do this as follows:
> d = Draw.MolDraw2DSVG(350,300,-1,-1,False)

Perfect, thanks! Maybe the keyword can be `useFreeType = False` or something 
along those lines.

-Geoff



___
Rdkit-discuss mailing list
Rdkit-discuss@lists.sourceforge.net
https://lists.sourceforge.net/lists/listinfo/rdkit-discuss

[Rdkit-discuss] SVG depiction with fonts?

[Geoffrey Hutchison]

Hi all,

I recently upgraded to RDKit 2021.3 from the March 2020 version. With last 
year's release, I was able to tweak the generated SVG depictions to replace 
characters (e.g., where we used "*" in a SMILES but really wanted "M" for a 
metal center) or change the font-weight and font-size.

svg.replace("font-weight:normal", "font-weight:bold")

Now it seems as if the characters are turned into strokes. Is there an option 
to turn this off and go back to SVG characters with font-weight attributes?

Thanks,
-Geoff

---
Prof. Geoffrey Hutchison
Department of Chemistry
University of Pittsburgh
tel: (412) 648-0492
email: geo...@pitt.edu
twitter: @ghutchis
web: https://hutchison.chem.pitt.edu/



___
Rdkit-discuss mailing list
Rdkit-discuss@lists.sourceforge.net
https://lists.sourceforge.net/lists/listinfo/rdkit-discuss

[Rdkit-discuss] Avogadro2 scripts for RDKit

[Geoffrey Hutchison]

Dear all,

For years, people have asked about running RDKit inside Avogadro (e.g., 
generating conformers with ETKDG instead of Open Babel).

The new Avogadro 1.94 release makes it much easier to download and install 
Python scripts - Avogadro hands off a SDF file, something happens and RDKit 
hands back a new SDF.

i started a few quick scripts:
https://github.com/ghutchis/avogadro-rdkit

- ETKDG v3 conformer
  (soon this will generate a specified number of conformers and rank with UFF 
or MMFF94 .. also will add all conformer coordinates)
- Generate canonical tautomer
- Optimize with UFF or MMFF94

The scripts are tiny but useful - and most is boilerplate to pop up dialogs or 
set the menu item.

I don't know how many people use Avogadro here, but I'm curious for suggestions 
on common "molecule building" tasks:
- Full standardize(?)
- Canonical conformer frame of reference?

Thanks,
-Geoff

___
Rdkit-discuss mailing list
Rdkit-discuss@lists.sourceforge.net
https://lists.sourceforge.net/lists/listinfo/rdkit-discuss

[Rdkit-discuss] RDKit - contributing conformational entropy descriptor

[Geoffrey Hutchison]

Hi Greg,

We just published a paper with a linear model predicting conformational 
entropies:
https://pubs.acs.org/doi/10.1021/acs.jctc.0c01213
https://github.com/hutchisonlab/molecular-entropies

We built the notebooks on top of RDKit - and of course would like to contribute 
the descriptor back to RDKit.

Is there a contribution guide (e.g., how to structure the code before a pull 
request)? In particular, this one has a few components that might prove useful 
as separate calls (e.g., the ring flexibility measure).

Thanks,
-Geoff

---
Prof. Geoffrey Hutchison
Department of Chemistry
University of Pittsburgh
tel: (412) 648-0492
email: geo...@pitt.edu
twitter: @ghutchis
web: https://hutchison.chem.pitt.edu/



___
Rdkit-discuss mailing list
Rdkit-discuss@lists.sourceforge.net
https://lists.sourceforge.net/lists/listinfo/rdkit-discuss

Re: [Rdkit-discuss] chair vs boat detection

[Geoffrey Hutchison]

Yes, you can detect chair / boat (etc.) using Cremer-Pople parameters 
(particularly the angles).

There is a bunch of related RDKit code from our recent ring-puckering paper:
https://doi.org/10.1021/acs.jcim.0c01144 
<https://doi.org/10.1021/acs.jcim.0c01144>
https://github.com/lucianlschan/RING

In particular - you want to call GetRingPuckerCoords(coordinates)
https://github.com/lucianlschan/RING/blob/master/Ring_Analysis.py

Hope that helps,
-Geoff

---
Prof. Geoffrey Hutchison
Department of Chemistry
University of Pittsburgh
tel: (412) 648-0492
email: geo...@pitt.edu
twitter: @ghutchis
web: https://hutchison.chem.pitt.edu/

> On Mar 12, 2021, at 1:10 PM, Paul Emsley  wrote:
> 
> 
> On 12/03/2021 05:11, Ling Chan wrote:
>> 
>> 
>> Just wonder if there is any function to distinguish between a chair ring and 
>> a boat ring?
>> 
>> Don't worry if there is no such utility. I can write my own geometry 
>> detection. Just that I don't want to reinvent the wheel.
>> 
> 
> Isn't this Cremer Pople?
> 
> http://enzyme13.bt.a.u-tokyo.ac.jp/CP/
> 
> https://smb.slac.stanford.edu/facilities/software/ccp4/html/privateer.html
> 
> 
> Paul.
> 
> 
> 
> 
> ___
> Rdkit-discuss mailing list
> Rdkit-discuss@lists.sourceforge.net
> https://lists.sourceforge.net/lists/listinfo/rdkit-discuss

___
Rdkit-discuss mailing list
Rdkit-discuss@lists.sourceforge.net
https://lists.sourceforge.net/lists/listinfo/rdkit-discuss

Re: [Rdkit-discuss] Looking for additional GSoC co-mentors

[Geoffrey Hutchison]

Wearing my "admin for Open Chemistry" hat for the moment.. As a community, 
we've been very lucky to get a lot of good open source chemistry development 
done over the last few years.

Moreover, many of these students are now exposed to RDKit, cheminformatics, and 
good coding. (In other words, they're good recruits for both academic and 
industrial positions. ;-)

Google is never open about their criteria for allocating the number of slots to 
organizations. Still, groups with more mentors are more likely to get more 
slots.

I'll be more blunt than Greg - without more mentors we will be forced to make 
very hard choices. We'll probably still need to do that, but finding a few more 
co-mentors will certainly help RDKit.

-Geoff


> On Apr 1, 2020, at 2:16 AM, Greg Landrum  wrote:
> 
> Dear all,
> 
> Yesterday was the last day for potential students to submit applications for 
> Google Summer of Code and we got proposals for the following five projects:
> 1) RDKit integration with MongoDB. Python based
> 2) Implement a generalized file reader and a multi-threaded file reader. C++ 
> based
> 3) Integrating trained neural networks (specifically ANI-like force fields) 
> into the RDKit. C++ based
> 4) RDKit-OpenMM integration. C++ based
> 5) Improved RDKit integration with Jupyter, Dask, Pandas, Plotly, and Bokeh. 
> Python based
> 
> There's more about the first four projects here: 
> https://wiki.openchemistry.org/GSoC_Ideas_2020#RDKit_Project_Ideas 
>  and I'm 
> happy to answer questions about them.
> 
> We have a mentor and co-mentor for each of these projects, but I'm looking 
> for a few more people to act as co-mentors. It doesn't make sense to do a 
> project without at least two mentors - one primary mentor and one co-mentor- 
> and Google is strongly encouraging organizations to have three mentors - one 
> primary and two co-mentors - available for each project. Being a co-mentor is 
> going to require an hour or two a week on average over the course of the 
> program.
> 
> There's more information on the program, including the dates, here:
> https://summerofcode.withgoogle.com/how-it-works/ 
> 
> 
> GSoC is really a great program and being a mentor/co-mentor is a good way to 
> help move both the open-source community and the RDKit forward.
> 
> If you're interested or have questions, feel free to send me email,
> -greg
> 
> ___
> Rdkit-discuss mailing list
> Rdkit-discuss@lists.sourceforge.net
> https://lists.sourceforge.net/lists/listinfo/rdkit-discuss

___
Rdkit-discuss mailing list
Rdkit-discuss@lists.sourceforge.net
https://lists.sourceforge.net/lists/listinfo/rdkit-discuss

Re: [Rdkit-discuss] RDKit Num Rotors descriptors?

[Geoffrey Hutchison]

> PS: I found that if you want to use the non-strict version, you can do this:
> 
> from rdkit.Chem import rdMolDescriptors
> rdMolDescriptors.CalcNumRotatableBonds(mol, 0)
> 
> (where 0 means NonStrict, 1 means Strict, and -1 means Default.)
> 
> I tested this with a molecule containing t-butyl and got different results 
> for strict and non-strict, as expected, and the default was the same as 
> strict.

Excellent, thanks!

-Geoff



___
Rdkit-discuss mailing list
Rdkit-discuss@lists.sourceforge.net
https://lists.sourceforge.net/lists/listinfo/rdkit-discuss

[Rdkit-discuss] RDKit Num Rotors descriptors?

[Geoffrey Hutchison]

I'm using a script in RDKit to grab a bunch of descriptors for QSAR / ML. I 
have a particular question about the # of rotatable bonds:

Consider:

from rdkit import Chem
from rdkit.Chem import Descriptors
example = Chem.MolFromSmiles("CN(C(=O)OC(C)(C)C)C(=S)NC(=O)OC(C)(C)C")
Descriptors.NumRotatableBonds(example)
0

example2 = Chem.MolFromSmiles("CC(C)(C)CCC(C)(C)C")
Descriptors.NumRotatableBonds(example2)
1


Can someone explain why a terminal butyl isn't a rotatable bond? Based on 
mailing list questions, I thought NumRotatableBonds was supposed to return 
matches for the SMARTS "[!$(*#*)&!D1]-&!@[!$(*#*)&!D1]"  (e.g. what's defined 
in the Lipinski code).

Thanks,
-Geoff



___
Rdkit-discuss mailing list
Rdkit-discuss@lists.sourceforge.net
https://lists.sourceforge.net/lists/listinfo/rdkit-discuss

Re: [Rdkit-discuss] molecular dynamics using RDkit only

[Geoffrey Hutchison]

> Even though it needs polishing and adding unit tests, this code is functional 
> and usable
> 

I expect that many people (myself included) would be happy to help with polish 
and unit tests for this. Many thanks Paolo!

-Geoff


___
Rdkit-discuss mailing list
Rdkit-discuss@lists.sourceforge.net
https://lists.sourceforge.net/lists/listinfo/rdkit-discuss

Re: [Rdkit-discuss] Open-source business models and the RDKit

[Geoffrey Hutchison]

> - That last bullet point likely works for academics too: think about adding 
> some support for open-source development to your next grant proposal. I would 
> assume that there are ways to engineer this.

I have the most experience with this point, naturally. Suffice to say that I've 
written support letters for numerous NIH and NSF-funded projects. In NSF lingo, 
open source development for widely-used packages is a clear "broader impact" 
and generally appreciated by reviewers.

The big challenge on the academic side is similar to what Andrew mentioned - 
that it's easier to get people (i.e., funding sources) interested in new 
feature X than in testing, bug fixing and documentation.

Greg's point about support contracts I think is important - there's a definite 
need for this and I hope to see this area grow.

Cheers,
-Geoff

___
Rdkit-discuss mailing list
Rdkit-discuss@lists.sourceforge.net
https://lists.sourceforge.net/lists/listinfo/rdkit-discuss

Re: [Rdkit-discuss] chemfp preprint

[Geoffrey Hutchison]

> Sometimes, I wish there was a rdkit consortium/NPO (so that donations are tax 
> deductible), so that rdkit could be massively funded by all its commercial 
> users, and even accepting individual donations.


I don't want to hijack the thread, so please feel free to take this off-list 
with anyone interested.

I think it's an interesting idea in general in open chemistry. We have set up 
an Open Chemistry collective - this receives $$ from Google Summer of Code. The 
"host" is the Open Source Collective, a 501c6 non-profit in the United States 
(https://docs.opencollective.com/help/hosts/open-source-collective)

The collective isn't perfect, it skims 5% for transaction fees and overhead, 
but it's:
- completely transparent for donations
- completely transparent for expenses
- allows both one-time and recurring donations

Greg can correct me - I think we handled the $$ to RDKit from Google Summer of 
Code 2018 before we set this up, but it's certainly there to use. You can 
create your own RDKit collective pretty easily too:
https://opencollective.com/open-chemistry

One big benefit is that OpenCollective handles all the legal paperwork and 
accounting.

-Geoff

PS One regret is that I haven't had need of chemfp in house, or I would have 
pushed some $$ towards Andrew.

___
Rdkit-discuss mailing list
Rdkit-discuss@lists.sourceforge.net
https://lists.sourceforge.net/lists/listinfo/rdkit-discuss

Re: [Rdkit-discuss] Evaluating ETKDG method with the Platinum Dataset

[Geoffrey Hutchison]

To give some context, it's not that we're trying to sample a diverse
set of conformers (and find something close to the experimental). In
this case, we're generating initial geometries - to assess Naruki's
fragment-based builder in Open Babel.

But you raise an excellent point - by picking only one random
conformer (as we're doing), we'll absolutely going to have a higher
RMSD than sampling 50 conformers per compound and picking the best.

We'll try a quick test to be safe, but thanks for the suggestion.

-Geoff

On Mon, Feb 25, 2019 at 3:55 PM Greg Landrum  wrote:
>
> Hi Naruki,
>
> You're only generating a single conformer per molecule; I wouldn't expect 
> that to do particularly well. It's generally better to call 
> EmbedMultipleConfs().
>
> As an aside: I've looked at the platinum set too, it might be worth checking 
> out this RDKit blog post: 
> http://rdkit.blogspot.com/2017/05/looking-at-platinum-dataset.html
>
> -greg
>
>
> On Mon, Feb 25, 2019 at 11:53 AM Naruki Yoshikawa 
>  wrote:
>>
>> Dear all,
>>
>> I'm evaluating ETKDG method implemented in RDKit using the Platinum
>> Dataset introduced in a benchmark paper
>> https://pubs.acs.org/doi/abs/10.1021/acs.jcim.7b00505/
>> SMILES generated from the dataset is served as input and a 3D
>> conformer is generated.
>> We evaluate RMSD between generated structure and experimental structure.
>>
>> Although the author of the benchmark paper reported the mean RMSD to
>> be below 1.0 angstrom, my evaluation code reports around 1.5 angstrom.
>> I can't figure out why such a big difference occurs.
>>
>> My evaluation code is here:
>> https://gist.github.com/n-yoshikawa/0ba04a1b0c718c4cc8d83702f3759afa
>> There is a link to data in this gist.
>>
>> Thanks,
>> Naruki
>>
>>
>> ___
>> Rdkit-discuss mailing list
>> Rdkit-discuss@lists.sourceforge.net
>> https://lists.sourceforge.net/lists/listinfo/rdkit-discuss


___
Rdkit-discuss mailing list
Rdkit-discuss@lists.sourceforge.net
https://lists.sourceforge.net/lists/listinfo/rdkit-discuss

Re: [Rdkit-discuss] issues with explicit / implicit valence

[Geoffrey Hutchison]

> Essentially I'm taking the result of a Gaussian optimization (for a radical); 
> constructing an SDF file with OpenBabel (via cclib), and then trying to read 
> the result in RDKit.
> I have the SMILES string of the radical, but the connectivity is lost in the 
> gaussian output file. So the SDF that gets created by OpenBabel has to assume 
> bond orders based on distances that it sometimes gets wrong.

As someone who uses similar workflows, my suggestion is to save the original 
SMILES or SDF file, and merely update the coordinates. That is, if you use OB 
or RDKit to generate an initial 3D structure, save a SDF with the bond valence 
information (and in this case, presumably the M RAD record). Then you read the 
Gaussian output with Open Babel or cclib.

Since the atom order should be the same from Gaussian as your original SDF, you 
merely have to copy over the coordinates after reading the Gaussian file.

You can, in principal, also save the SMILES string as the "title" line in a 
Gaussian or other quantum chemistry program if you don't want to keep duplicate 
files (i.e., when reading the output with Pybel you can use mol.title to get 
the SMILES string back).

Suffice to say the main thing is to keep the original bond/valence information, 
don't throw it away. ;-)

Cheers,
-Geoff

---
Prof. Geoffrey Hutchison
Department of Chemistry
University of Pittsburgh
tel: (412) 648-0492
email: geo...@pitt.edu
twitter: @ghutchis
web: https://hutchison.chem.pitt.edu/



___
Rdkit-discuss mailing list
Rdkit-discuss@lists.sourceforge.net
https://lists.sourceforge.net/lists/listinfo/rdkit-discuss

Re: [Rdkit-discuss] Can't import Chem from rdkit in Anaconda Python 3.6.5

[Geoffrey Hutchison]

>> Note that my answer assumes that there is a reason that you don't have X11 
>> installed on your linux box. If that's not the case, you should be able to 
>> fix things "more easily" by installing X
> 
> Quite frankly, this is rapidly becoming unusable as a software platform. I 
> need to install X11 to UUF-optimize a MOL? Seriously?

No, you can compile RDKit yourself if you don't want to use X11 features. You 
wanted to install through conda, which has a set of packages for 'most use' - 
YMMV.
(We have a version of RDKit on our server w/o X11)

My $0.02,
-Geoff
--
Check out the vibrant tech community on one of the world's most
engaging tech sites, Slashdot.org! http://sdm.link/slashdot
___
Rdkit-discuss mailing list
Rdkit-discuss@lists.sourceforge.net
https://lists.sourceforge.net/lists/listinfo/rdkit-discuss

Re: [Rdkit-discuss] looking for suggestions: github vs bitbucket vs google code

[Geoffrey Hutchison]

 [1] I believe it should be possible to continue to keep the svn
 repository in sync with the git code base in the same way that the
 current github RDKit version is kept in sync with the sf.net svn repo.

My $0.02 on this -- since I’m migrating Open Babel from SVN at sf.net to 
GitHub. GitHub offers very nice “transparent” SVN access:
https://github.com/blog/1178-collaborating-on-github-with-subversion

Hope that helps,
-Geoff
--
Everyone hates slow websites. So do we.
Make your web apps faster with AppDynamics
Download AppDynamics Lite for free today:
http://p.sf.net/sfu/appdyn_d2d_jan
___
Rdkit-discuss mailing list
Rdkit-discuss@lists.sourceforge.net
https://lists.sourceforge.net/lists/listinfo/rdkit-discuss

Re: [Rdkit-discuss] retrosynthesis

[Geoffrey Hutchison]

On Jul 23, 2012, at 9:32 AM, Greg Landrum wrote:

 I'm not aware of anything. The RDKit has many of the pieces necessary
 to start to build such a system, but a library of reactions to use in
 the retrosynthesis is missing. As I commented on your feature request

Indeed, several people have approached me (through Open Babel) with a similar 
request. As Greg said, there is no existing open database of reactions.

I've tried to catalyze the issue by asking Igor Filippov (of OSRA) to do 
reaction recognition (now in beta) and we've added reaction support for 
ChemDraw CDX files. This would help people compile such a reaction database 
from existing files and papers.

I also know that Abe Heifets at Toronto has been working on the code side of 
things, but he currently uses a commercial reaction database:
http://www.cs.toronto.edu/~aheifets/ChemicalPlanning/

In short -- it's a key problem, but hopefully can be solved through a bit of 
common work.

-Geoff
--
Live Security Virtual Conference
Exclusive live event will cover all the ways today's security and 
threat landscape has changed and how IT managers can respond. Discussions 
will include endpoint security, mobile security and the latest in malware 
threats. http://www.accelacomm.com/jaw/sfrnl04242012/114/50122263/
___
Rdkit-discuss mailing list
Rdkit-discuss@lists.sourceforge.net
https://lists.sourceforge.net/lists/listinfo/rdkit-discuss

Re: [Rdkit-discuss] PDB processing

[Geoffrey Hutchison]

On May 10, 2011, at 2:37 AM, Adrian Schreyer wrote:

 Open Babel can read PDB files but the last time I checked it was not possible 
 to get all the structure factors from atoms.

If someone can explain to me what you actually want, that can get fixed. (I'm 
more a materials scientist than a biochemist.) That's more of a subject for the 
Open Babel tracker` or mailing list.

And I completely understand Greg's view on processing PDB. It's one thing if 
you're talking about authentic PDB files, but the file format has been abused 
horribly and the OB version is a (necessary) mess. Any time we try to clean it 
up, people report a dozen bugs on their not-quite-standard files.

Cheers,
-Geoff
--
Achieve unprecedented app performance and reliability
What every C/C++ and Fortran developer should know.
Learn how Intel has extended the reach of its next-generation tools
to help boost performance applications - inlcuding clusters.
http://p.sf.net/sfu/intel-dev2devmay
___
Rdkit-discuss mailing list
Rdkit-discuss@lists.sourceforge.net
https://lists.sourceforge.net/lists/listinfo/rdkit-discuss

Re: [Rdkit-discuss] Bond-type perception.

[Geoffrey Hutchison]

On Apr 4, 2011, at 11:28 PM, Greg Landrum wrote:

 aromaticity, the RDKit does not do any bond perception; it assumes
 that bond types are set properly in the input molecules. So if you
 want to build methane, you need to set the four C-H bonds to type

Moreover, I'd be happy to donate bond perception code to RDKit, but these 
things only really work well when you have 3D coordinates. (I've tried many 
implementations.)

But I can't quite figure out how you'd want to perceive bond types without 
any obvious information.

Cheers,
-Geoff
--
Xperia(TM) PLAY
It's a major breakthrough. An authentic gaming
smartphone on the nation's most reliable network.
And it wants your games.
http://p.sf.net/sfu/verizon-sfdev
___
Rdkit-discuss mailing list
Rdkit-discuss@lists.sourceforge.net
https://lists.sourceforge.net/lists/listinfo/rdkit-discuss

Re: [Rdkit-discuss] beta of Q1 2010 release up

[Geoffrey Hutchison]

On Apr 19, 2010, at 11:59 PM, Greg Landrum wrote:

 The only issue seems to be that I don't know how to install the Python files 
 into my system site-packages. I just don't have the cmake experience.
 
 This is something I haven't looked into yet. I believe it should be
 possible to get everything into site-packages (and perhaps set RDBASE
 automatically in an appropriate manner), but I still need to figure
 out how to do it. Another possibility would be to do the python
 installation using distutils. This has the advantage of being more
 pythonic, allowing integration with pypi, etc.

Feel free to steal the CMake from Open Babel. See scripts/CMakeLists.txt. A 
few weeks ago, we used CMake's SWIG support to compile and install the Python 
bindings -- this was buggy.

Now, we run distutils, e.g.:
install(CODE execute_process(COMMAND ${PYTHON_EXECUTABLE} 
${openbabel_SOURCE_DIR}/scripts/python/setup.py install 
--prefix=${CMAKE_INSTALL_PREFIX} WORKING_DIRECTORY 
${openbabel_BINARY_DIR}/scripts))

There's actually a Win32 version, and a UNIX version. The Mac version would be 
something like:
install(CODE execute_process(COMMAND ${PYTHON_EXECUTABLE} 
${openbabel_SOURCE_DIR}/scripts/python/setup.py install WORKING_DIRECTORY 
${openbabel_BINARY_DIR}/scripts))

Hope that helps,
-Geoff
--
Download Intel#174; Parallel Studio Eval
Try the new software tools for yourself. Speed compiling, find bugs
proactively, and fine-tune applications for parallel performance.
See why Intel Parallel Studio got high marks during beta.
http://p.sf.net/sfu/intel-sw-dev
___
Rdkit-discuss mailing list
Rdkit-discuss@lists.sourceforge.net
https://lists.sourceforge.net/lists/listinfo/rdkit-discuss

Re: [Rdkit-discuss] beta of Q1 2010 release up

[Geoffrey Hutchison]

On Apr 20, 2010, at 2:04 PM, Andrew Dalke wrote:

 In trying to get OpenBabel to compile on my Mac last week, working with Noel 
 on this, I found that this part of the OB configuration for the Mac doesn't 
 work quite right.
 I expect the default installation to go into the standard Python 
 site-packages directory, which for me is
 
  /Library/Python/2.6/site-packages

Yes, I need to fix Noel's CMakeLists -- it needs a special case for if(APPLE) 
here. That should show up in SVN in the next 15 minutes or so.

 I don't have a good feeling for how setting CMAKE_INSTALL_PREFIX should 
 affect the install. Perhaps there should be a CMAKE_INSTALL_PYTHON_PREFIX or 
 something like that. My observation here is only that when nothing is 
 specified then the default Python installation directory should be the 
 default determined by Python, and not by the cmake configuration.

Well, I haven't tried, it, but if you can give me a python script which gives 
the appropriate destination directory, I can hack the CMake code. e.g.

EXECUTE_PROCESS(COMMAND ${PYTHON_EXECUTABLE} -c 
import numpy; print numpy.get_include()
OUTPUT_VARIABLE NUMPY_INCLUDE_DIR)

This particular bit gives CMake the location of the numpy headers. I'm sure 
there's some similar function call for distutils, right?

Cheers,
-Geoff
--
___
Rdkit-discuss mailing list
Rdkit-discuss@lists.sourceforge.net
https://lists.sourceforge.net/lists/listinfo/rdkit-discuss