[Top][All Lists]

[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]

Re: [Bkchem-user] OASA, missing atoms in periodic_table.py

From: Loic d'Anterroches
Subject: Re: [Bkchem-user] OASA, missing atoms in periodic_table.py
Date: Wed, 07 Apr 2010 09:18:02 +0200
User-agent: Thunderbird (X11/20100317)


next time I will not put 2 subjects in one email. Sorry for the hashed
thread it creates...

Noel O'Boyle wrote:
> On 6 April 2010 13:56, Beda Kosata <address@hidden> wrote:
>> Loic d'Anterroches wrote:
>>> OC wrote:
>>>> Loic d'Anterroches wrote:
>>>>> just a small request, Neodymium[1] is missing from periodic_table.py and
>>>>> I do not understand the format of the periodic_table dictionnary to add
>>>>> it. If I can get some details about the "en", "els" and "valency" keys,
>>>>> then I can add the missing components. I have a database of about 50k
>>>>> molecules and when drawing them, the molecules with atoms not in the
>>>>> periodic_table are failing.
>>>> I'm not a chemist, but I bet it is:
>>>> "en" : electronegativity (Pauling). I notice that some values are not
>>>> exact (for example H 2.10 instead of 2.20) but it may be on purpose.
>>>> "els" : number of electrons in the neutral atom on the outer shell
>>>> "valency" : possible number of bonds
>>> Thanks, this was this valency where I was wondering the most. The
>>> numbers are all listed in ascending order, except for Carbone with 4
>>> first and 2.
>>> I will simply try and see what is coming out of it.
>>> Thanks!
>>> loïc
>> Hi,
>> valency is sorted from the most common to the least common (see for
>> example "P").
>> A complete periodic table is part of the development version of OASA.
>> You can either get it from the tla sources
>> (http://bkchem.zirael.org/arch_en.html) or it was distributed with the
>> last preview version of BKChem (a separate release of OASA will be made
>> for the final version).
>> The algorithm for 2D layout is indeed not optimal as it does not take
>> other that local influences into account and thus allows overlap of
>> parts of the structure. AFAIK CDK
>> (http://sourceforge.net/apps/mediawiki/cdk/index.php?title=Main_Page)
>> contains a more sophisticated algorithm, but I did not try it myself.
> Regarding structure diagram generation, please see
> http://baoilleach.blogspot.com/2008/10/cheminformatics-toolkit-face-off.html
> and 
> http://blueobelisk.shapado.com/questions/2d-structure-generation-background-and-techniques.

Thank you both of you for the answers, I have bookmarked everything and
downloaded the code, I will dig dipper into all that and let you know
what comes out.


reply via email to

[Prev in Thread] Current Thread [Next in Thread]