Using Invariom Tool

Here You can talk about MoleCoolQt
Link: MoleCoolQt Homepage

Moderator: chuebsch

IssyKirby
Posts: 1
Joined: 23 Apr 2012, 13:18

Using Invariom Tool

Post by IssyKirby » 17 Jun 2013, 14:22

I am using MoleCoolQT to assing Invarioms to atoms with a structure. However I am having problems when it comes to transferring these Invarioms and the multipole populations and other parameters from the database.
Has anyone else seen, have examples of how to do this?


Thanks,

Issy

tulenik
Rotating Anode With Optics
Posts: 336
Joined: 28 Nov 2010, 08:13
Location: Bratislava, Slovak Republic

Re: Using Invariom Tool

Post by tulenik » 22 Feb 2015, 10:54

What type of problem(s) you have exactly?
Aren't some invarioms missing in the database? Such invarioms are printed in red in the status window. With some luck you can find other suitable invariom (but it will be not so good fit, understandably) in the database (it is plain text file).
In such case the export to XD and/or MoPro files is disabled and you'll obtain only (if I can recall it exactly) something like invariom.in file in the folder with your structure of interest with the list of invariomes obtained from the structure.
If this is an issue I had the same "problem" tomorrow ;) however taking into account that latest version of invariom database is more than 400 days old (according to what MoleCoolQt wrote after database download) and for example latest version of UBDB database is almost 3 years old also [however it is somewhat questionable - if you find the original version in Buffalo, it is from 2006 - however after some searching you should also find 2012 version and updated lsdb utility for database transfer, capable of transfer not only to XD but also to MoPro and I am really not sure if there isn't any even more current version somewhere], I am afraid that only possibility how to handle such cases is to compute missing pseudoatoms by self... taking into account my almost zero background in theoretical computations (however I have "on board" some people able to do that and refinement against theoretical structure factor wouldn't be a problem for me) and absence of any "how-to for dummies" it is almost hopeless waiting for the miracle :|

My advice is to try also UBDB database, I think with my "problematic" case (see the picture - even for such simple structure it should be problematic obtaining all pseudoatoms needed - exactly because of this I think there is your problem) it had less missing pseudoatoms like in the case of invariom database. Ironically, ELMAM2 and 3beta databases have only 1 pseudoatom missing.
iran.png
As you can see now, whole matter is not so simple as it could be. I am trying to adopt this method for our problems, but the main problem is total lack of documentation at the tutorial level. It is somewhat better for XD, but I am using MoPro and Jana2006 (yes it is even possible using pseudoatoms in Jana - after brutal manual editing) and especially MoPro documentation is almost none. I found these two PhD thesis helpful:
https://depotuw.ceon.pl/browse?value=ch ... pe=subject
especially Katarzyna's thesis was helpful as an introduction to the topic. Even "self-evident" input file for lsdb was not so self-evident without Katarzyna's explanation :mrgreen: (use:

Code: Select all

lsdb -h > example.inp
to generate the file, according to - again - extremely brief instruction sole

Code: Select all

lsdb -h
is not generating anything :lol: )

Some tips or feature requests for Christian & co.:
1. built-in warning system sending requests about missing & wanted pseudoatoms to ready-and-willing database maintaners/developers :lol:
2. including UBDB transfer tool
You do not have the required permissions to view the files attached to this post.
Last edited by tulenik on 23 Feb 2015, 03:34, edited 2 times in total.

tulenik
Rotating Anode With Optics
Posts: 336
Joined: 28 Nov 2010, 08:13
Location: Bratislava, Slovak Republic

Re: Using Invariom Tool

Post by tulenik » 22 Feb 2015, 15:56

Hm, now I found that downloading database from Internet is possibly broken. And the database I found for manual download is from 2008... maybe your problem is because of temporarily impossible download?
My question is, what is downloading address for latest version of the database :?: evidently what I found here:
http://ewald.ac.chemie.uni-goettingen.de/programs.html
is not the latest version

User avatar
chuebsch
300th member
Posts: 85
Joined: 10 Nov 2009, 18:56
Location: Bayreuth

Re: Using Invariom Tool

Post by chuebsch » 23 Feb 2015, 08:36

There was an error in the server that is why the download in MoleCoolQt was not working. So it should work again now.

tulenik
Rotating Anode With Optics
Posts: 336
Joined: 28 Nov 2010, 08:13
Location: Bratislava, Slovak Republic

Re: Using Invariom Tool

Post by tulenik » 24 Feb 2015, 10:08

Christian, slight off-topic: is here some tutorial for dummies how to compute missing invarioms? Provided I can obtain access to supercomputing center (I don't suppose it is reasonably fast on common PC hardware) with at least Gaussian and Tonto and I am able to do multipole refinement in Jana2006 (and I suppose in MoPro also). When I am talking about dummies I must admit I am - at least for now - totally inept about theoretical calculations, but I have "top grade" theoretician on board (our boss).

Return to “MoleCoolQt”