HCI and ChemViz
After a long hiatus, I'm back again.
Here at CSCS we are planning the development of a Molecular Structure Editor. There are so few free ones and our users ask for specific functionalities not found even in commercial ones.
But before starting we need to find funding and people, so I have time to study HCI (Human-Computer Interaction).
My plan is:
- Use this blog to collect good/bad HCI ideas from existing Chemistry Visualization tools
- Start from introductory HCI texts to have a foundation of the field
- Prototype, prototype...
Labels: GUI, HCI, visualization tools
posted by mario @ 07:14
4 comments
4 Comments:
2D or 3D?
There are indeed a few free ones. Even opensource ones... why not hook on to those projects, instead of reinventing the wheel?
3D
You are true about not reinventing the wheel...
But till now we have not found anything suitable. Please send any pointer you have about 3D modeling tools. The majority of my users use Molden for modeling. I have tried Gabedit and another one Windows only for which I forgor the name.
There is another reason for trying the "reinvent the wheel" route. We are pursuing a research project about representation in chemistry. Currently the project is based on a commercial visualization environment (AVS/Express) that unfortunately limits the possibilities of deployment and cooperation. So the modeler structure could be the base for backing this project.
Anyway, noting has been firmly decided till now.
Thanks for your comments
mario
There has been a lot of talk about the 'sculpt' mode; to add that to Jmol. In this mode, except for the basic adding of atoms, the user can drag atoms around, rotate bonds, etc, while the model is under the counter-force and/or guidance of a force that is being run in the background. A MM is used for that.
Bioclipse would be the best platform for this: it already has Jmol integrated (latest version), and Bioclipse has integrated MM/QM tools in the past, such as Ghemical. No one, however, has had the resources to make it happen. Additionally, Bioclipse integrates many other chemoinformatics tools, and inherits a lot of functionality from the Eclipse RCP. Furthermore, efforts are ongoing to add support for R (statistics), Subversion (collaborative working), scripting languages, etc. Since the paper in BMC Bioinformatics last year, a lot more funding has become available, and the developer and user communities have grown steadily, both in terms of number of people, as well as number of groups and countries.
There is a lot of code to reuse, and all the components seem already present (visualization, chemoinformatics tools, and MM/QM).
If you like more information, please contact me by email. I would happy to provide more details and/or existing code in action.
Egon always seems to be ahead of me in finding interesting blog articles...
I've been working on Avogadro for a while, which I think improves on Gabedit in terms of usability, if not feature list. One key task is that we have not yet finished isosurfaces, although that should only be another week or two.
I don't think Bioclipse (or Java in general) is very good for 3D molecular editing -- it's not fast enough. But it also depends on your needs. I work with small molecules and computational chemistry. So I need crystal structures, but not necessarily every field in a PDB file.
Mario... I'd be interested in your thoughts on the subject. I was also pondering this subject on my blog.
Post a Comment
<< Home