ConcreteChemViz

3DMolSym

2008-01-06T07:25:00.000+01:00

3DMolSym
http://www.molwave.com/software/3dmolsym/3dmolsym.htm
3DMolSym is an educational program designed to visualize the symmetry elements of molecules and to animate the corresponding symmetry operations in an interactive 3D environment.

HCI things I like

The rotate controls update a small icon to show the axis around which it rotate
Animation usage to explain symmetry
Captured view with a simple click (the button then becomes the view selection button)
The technology: it is a Shockwave web application.
The usage of transparent elements to show movement, symmetry planes

What I do not like

Captured view has a reset button that does not do the intuitive thing i.e. resetting the saved view.
The small view for panning on the left is a duplicate of the main window useful only because the main window has no pan method (e.g. the three mouse button function for rotate/zoom/pan)

HCI and ChemViz

2008-01-06T07:14:00.000+01:00

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...

Grid vs. Visualization conferences

2007-10-03T06:05:00.000+02:00

I'm just back from the first day of the EGEE'07 conference. I'm not a Grid fan, but I noted some differences between this and the visualization conferences I'm used to attend.
At the Grid conference I found, respect to a visualization one:

more Italian participants
more women
more gadgets available at the exhibition
more food and more beer and wine offered

Well, it is not a scientific comparison. It is also biased by the location (mostly US the visualization conferences, instead this was in Hungary). But the differences attracted my attention.
The reason for the differences could be:

More money from industry, they see commercial opportunities
Young field
In Italy we have discovered an empty field in which we are not behind the other countries, sadly
(my biased opinion) more vaporware...

Practical chemistry visualization at ETH Zurich

2007-08-31T13:48:00.000+02:00

Next week I will teach the short tutorial Practical introduction to chemistry visualization at ETH Zurich.

The idea is to transmit usable knowledge on visualization that could help chemists research work and help them consider more visualization in their research work.

I will appreciate if you can leave here some detail on your visualization experience and, after the course, comments/suggestions/critiques on it.

See you Monday!

Second Life for science

2007-07-04T12:43:00.001+02:00

Yesterday we had also an extensive demo of Second Life, the virtual world. Karl Harrison from Oxford University had built the "Oxford island" inside the Second Life world, to test how this concept could help science teaching. I do not understand how this concept could help, but surely the world has an interesting interaction model.

Similar experience a grandmother participant to the conference that meets her grandchild in a virtual world park.

Run, run, GRC conference!

2007-07-04T11:46:00.000+02:00

Two conference days passed so fast!

Scientific sessions and the poster area are full of interesting themes: from brain studies of perception to what chemistry students have in their minds. I like this interdisciplinarity. I like to mix ideas not only from the scientific visualization area. Maybe some new idea spark in the mind, maybe not. In any case it is enriching me.

One nice idea is to start the question time with five minutes of (forced? suggested? encouraged?) discussion with your seat neighbors. Yesterday the discussion continued also after the session in front of a beer.

But there are also 'undercover' activities: Monday an introduction to Flash; yesterday a Jmol tutorial and a GeoWall presentation.

Flash: I do not know if could be of interest to us, but it is really simple to create and animated 2D application. Various professors use it to explain concepts or to make interactive demonstrations.

Jmol: with the principal developer teaching, nothing less! Latest version rivals standard chem viz tools.

GeoWall: cheap alternative for stereo projection screens. It uses polarized glasses to separate L and R info.

Conference starts

2007-07-02T12:37:00.000+02:00

Seems the first day at the college: old friends reunions, laughs, ...

Before the two after-dinner sessions, an official from the GRC organization remembered (strongly remembered) the rules: no public relation of conference content.

Interesting rule: the rationale is to remove every possible block that could hinder the free discussion of advanced topics or new research lines. I agree, but this means here I can only post my sensations and the (public) titles of the talks.

So the two first sessions:

On visual perception: half known phenomenas, half a new effect that elicited a general "Wow!" from the audience.
On quaternions. Known mathematics, but some nice applications. Same of them were like an illusionist show.

Was hard to remain awake. But an interesting experiment was attempted: before questions, five minutes the people were forced to discuss together on the talk content. Nice method to make connections.

Design workshop ended

2007-07-01T23:21:00.000+02:00

Last half day for the Visualization Design Workshop at the Gordon Research Conference. The good things always finish...

The final exercise of the workshop presented four ideas on how to redesign a visualization. Here are some comments on mine. But the best visualization was the animation of a protein docking made for illustration, not exploration. Amazing!

Then Barbara Tversky spoke on External Representations and especially on Animation usage.

Interesting:

Animation is hard to perceive
Animations are conceived as discrete steps
It is showing, not explaining

That means, that I should reconsider how molecular animations are made. Currently they are simply the translation into images of the data received. Why I'm not able to extract important points? Why the sequence is never segmented into "worth note" and "transition" parts.

As usual the available time is the main constrain together with the difficulty to understand what the user want to demonstrate with the movie.

After dinner, the real Gordon Research Conference begins. Nice for the ones of us that are still not synchronized with the time zone...

GRC - The campus

2007-07-01T11:27:00.000+02:00

The conference is in the Bryant University campus. For someone not used at the American campuses, it is really impressive: it is new and it is rich. The university has only financial and economic faculties. And that explains everything...

The student dormitory instead are quite poor. An educational method or what?

But the seminary room, has incredibly comfortable chairs and big desks. It helps a lot the attention.

Not to say, wireless everywhere. In the afternoon I want to try under the trees.

As a student, would I want to stay here? I don't know. It is far from everywhere. Seems to me you are forced to study a lot.

The Design Workshop

2007-07-01T10:52:00.000+02:00

Before the Gordon Research Conference start I'm attending the pre-conference workshop on Design Principles for Creating Effective Visualizations.

It is short lectures and group work. A lot of group work!

Here are some sparse notes:

Visualization is the use of graphical techniques to convey info and support reasoning
Representation: The Mike Hann's "Ceci n'est pas une molecule". We are showing something that highlight some aspect of the original data, but it is not the original data. I have to read the handout form a Donald Norman book (Things that make us smart).
The enormous power of representation that can change significantly the cost of the insight.
Distributed cognition: it is the internal representation plus the external cognition artifacts that play together
Visualization challenge: Design a representation that uses graphical techniques effectively. Design needs three interlocked disciplines: psychology, domain knowledge and art&design. The last is always forgotten.
Then divided in 4 groups we selected good/bad visualizations. I presented the Ute Roehrig Alzhaimer proteins aggregation and the Davide Branduardi energy landscape (see my workshop materials page)
Constructive criticisms: Do the audience knows what they are going to see? Challenge of conventions. The landscape is good because matches everyday experience (valleys, things going down).
From another work (movement of lipids in a cellular membrane): use of highlight to direct attention. Animation of H atoms confusing, need same form of traces to be understandable. Good idea the use of zoom to attract attention.
Unfortunately not open, but on flashchem.nelson.com there are nice examples of interactive Flash usage for concept explanation.
Barbara presented idea on cognitive design principles. The viz designer must uncover the mental model of the viz user. Use the context to disambiguate the message.
Color usage ideas: known guidelines. But a new ideas: hue for periodic data. Hue is circular, so can convey ideas on things like phase (that it is circular). It also show very clearly discontinuities.
Space usage: look at cartography. To show one thing better, the cartographer must absolutely distort something else.
The viz scientists learn by observing. Learn by examples.
Design a visualization. Do not lock into a solution too early!

Today we must present a design to make more effective one presented visualization. More on this later. But seems we are not able to select one visualization...

The Gordon Research Conference

2007-07-01T10:31:00.000+02:00

I'm near Providence RI for the Gordon Research Conference on Visualization in Science and Education.

Do you know what are those conferences? They are not usual conferences with publications, proceedings and so on. The main idea is to join people from diverse fields but with a common interest to freely discuss and exchange ideas. For this reason it is strictly prohibited to make any reference to work presented at the conference.

To quote from the home page:

The Gordon Research Conferences promote discussions and the free exchange of ideas at the research frontiers of the biological, chemical and physical sciences. Scientists with common professional interests come together for a full week of intense discussion and examination of the most advanced aspects of their field. These Conferences provide a valuable means of disseminating information and ideas in a way that cannot be achieved through the usual channels of communication - publications and presentations at large scientific meetings.

My impressions till now:

The atmosphere is the one of a college or school, not of a conference. No one seems urged to demonstrate anything
Multidisciplinarity at the top! Psychologists, chemistry professors, educators, a documentary maker, biologists...
Real interest for each other work
Close contact with famous names in the visualization area: Pat Hanrahan, Barbara Tversky

Now I'm in the middle of a pre-conference workshop on Design Principles for Creating Effective Visualizations. More on this later.

Molecular Visualization Day - my talks

2007-04-30T11:38:00.000+02:00

Beyond Ball-and-Stick – Representation and Perception in Chemistry Visualization

The morning talk was about:

Representational conformism in chemistry visualization software
How a good representation could enhance understanding
How all the chemistry representations are metaphorical
How standard representation could limit insight
All this with examples of strange, new representations and how standard representations could be chosend with different criterias.

From Tradition to Insight Support – What we can demand from our Chemistry Visualization Tools

The final talk was on tools. I presented my tools survey that covers:

Usability problem with visualization tools
Divorce between the chemists' mental process and what a tool offers.
And how visualization tool are mostly used for presentation and not for discovery

Ideas and questions from the audience:

As an example I cite Photoshop as one example of perfect tools in its field. Someone said that it is considered perfect because it is the bestseller of its category. Chicken and egg problem... anyway, the best seller draw the path and must stay in the top. That means it is forced to be perfect every time more in my opinion.
Is there any universal data file format? Yes! It is called CML (Chemystry Markup Language) but unfortunately no one care about it.
Why instead of doing more, the tools try to do less? That means, why the tools does not try to be built around components? Good suggestion. One example is OpenBabel. It tries to create a component devoted to data file input/output and nothing more. Good idea, but has shortcomings.
Visualization for the disabled. In UK if a blind person asks to take a visualization course you are forced to take him. I never had this kind of problem, but nonetheless it is something worth thinking about. For example studying auralization or considering broadly visualization as perceptualization.
How do you show something that is missing? (from the Stefano Leoni talk). For example it is difficult to show a vacancy or where water transforms into vapor.
An image is worth 10.000 words, but should be the right 10.000 words!

Molecular Visualization Day - talks

2007-04-30T10:55:00.000+02:00

Stefano Leoni, Max-Planck Institute for Chemical Physics of Solids
Thinking with the Eyes: Productive Visualization in Natural Sciences

Real life examples of visualizations from his own research. A lot of images were done with my STM4, but he added an interesting creative touch. Notable highlights:

Illuminated streamlines to show atom movements
Translucent polyhedra to draw attention to the important parts of the scene
Busy ball-and-stick structure where, due to the crossing of bonds, the formation of a phase transition seed was highligthed

Matt Cooper, Department of Science and Technology (ITN), University of Linköping
Evaluating the Effectiveness of Haptics for Biomolecular Education

Interesting study on the effectiveness of haptic interfaces for a protein docking problem. He made a distinction between been used to an interface method and having help from the interface. At the end the conclusion was thatreally haptic does not help much for understanding of data. Another consideration worth noting is that the haptic force feedback device needed was not a simple 3D one, but a 6D one (position plus rotations). That means higher cost and computation requirements.

He also defined visualization as a process with an image at the end. Later someone corrected the definition to: visualization is a process with an image in the middle. Quite true, also if compared to the James Watson discovery loop I show in my talks.

I envy the T-shirt Matt was wearing: the capsicine structural formula on the front and hot peppers on the back!

Jens Thomas, Computational Chemistry Group, Daresbury Laboratory
CCP1GUI Visualization System

A GUI for the Gamess-UK computational code. Besides the usual structure viewer, it featured an interesting Grid interface for the computational code. The idea was to create and unified, simple interface for computationale chemists so they have a single entry point for all their works. Interesting. The code is written in Python and is available.

Ganga Periyasamy, School of Chemistry, University of Manchester
Visualization to understand magnetically induced currents in transition metal complexes

This talk was interesting because visualization were disappointing...

She spoke on her research about the electronic currents in various compounds. The computations were done with Matlab that produced the 2D contour charts (1 a.u. below, on the molecule plane and 1 a.u. above). Due to the scarcity of points, Matlab produced horrible artefacts instead of smooth contours of the modulus of the vector currents).

This problem has been already solved (see the work of Daniel Sebastiani here at CSCS), but in my opinion the interest of this talk derived from a study of the problems a non-specialized tool could create for visualization.

Manchester Molecular Visualization Day

2007-04-30T10:32:00.000+02:00

Last April 20 the University of Manchester organized a Molecular Visualization Day to advertise what visualization could do to help chemistry researchers.

I was the star of the show with two talks (here are the abstracts and slides). I tried to cover my research on chemistry representations and visualization tools. The questions were stimulating (and the hospitality splendid too!).

Next posts I will cover ideas from the other speakers and comments and ideas from my own talks.

STM4 released!

2007-04-27T11:39:00.000+02:00

At last I have released the new varsion of my chemistry visualization environment STM3. It is obviously called STM4.

If you already know STM4, take a look to the new STM4 pages.

If you do not know what STM3 is look at its pages. Briefly STM3 does not want to sobstitute any existing chemistry visualization tools, but istead provides an environment in which to implement unusual and advanced visualization techniques as requested by my users.

STM4 is built on top of the commercial visualization development environment AVS/Express. This is good (it saves me from developing everything needed, it provides visualization techniques not normally available in a chemistry visualization tool) and bad (you need to buy a license).

Currently STM4 has been tested and compiled under:

Linux Suse 10.x
Windows XP
Mac OSX 10.4
Linux Red Hat Enterprise Linux AS release 4 (64 bits)

PDB heuristics

2007-02-17T09:28:00.000+01:00

The PDB file format has many advantages, it is well documented, it permits description of many different kind of information... But has also a lot of shortcomings, not least the fact that you need heuristics to extract the real structure stored inside.

I just stumped over the paper: PDB: Cruft to Content that collect a long list of heuristics needed to extract the real structure descibed by some PDB file avoiding the errors and inconsistencies present in many files deposited in the Protein Data Bank.

As soon as possible, I will implement some of them in my STM3 PDB file reader.

Documented file formats

2007-02-05T17:33:00.000+01:00

Where is the ultimate documentation on chemistry file formats? Here is what I have found

MOL2 - It is from a commercial company so it is well documented
Gaussian cube - There is a note in the VMD documentation that change a little the meaning of some info contained in the file
XYZ - again the molfile VMD page is a gold mine of file formats descriptions
MOL - Well documented
PDB - Well documented
PDB-Q - An old version of PDB, prior to 1992. Should be documented somewere...
CHGCAR - The insight about the format arrives not from the usual VMD page but directly reading the VMD reader plugin code
DCD - I remember a difference between what has been described and the real file format
DL_POLY The HISTORY file is well documented in the manual
Gulp
POSCAR
Shel-X
Siesta - Seems no more active
XDATCAR

Misplaced creativity in file formats

2007-02-02T04:50:00.000+01:00

Just finished revamping my chemistry data readers for STM4. Ang again I'm thinking about why a lot of creativity is misplaced in inventing new, similar un-parseable file formats!

Here are the formats STM4 supports for now. After the list I try to collect some of the nice (groan!) features of each of them. Why chemistry programs developers do not try to converge on a single format?

CHGCAR - VASP format containing also volume data
Gaussian cube - Gaussian containing also volume data
DCD - FMD, add to PDB trajectory data
DL_POLY - DL_POLY HISTORY file
FpStudio - FullProf Suite
Gulp - GULP input file
MOL - From MDL
MOL2 - From Tripos
PDB - Protein Data Bank
PDB-Q - An old version of PDB
POSCAR - VASP, also as concatenated file
SHEL-X - Shel-X crystallography program
Siesta - Siesta
XDATCAR - Another VASP animated format
XYZ - The simple xyz format, animated also
XYZ plus unit cell - idem plus a supporting file containing the unit cell

And now some of the complains:

CHGCAR - Could contain one or two sets of volumetric data. Why should be so difficult and unreliable to find the start of the second block? And why a division of the values by the cell volume based on the file name? And why no sensible extension to the file? And no atom type in the file.
Gaussian cube - Measurement units: Angstrom or Bohr?
DCD - It is a binary format, so almost works
FpStudio - Why two structures in the same file with different methods to describe symmetry? Why contains rendering options mixed with structural options?
Gulp - It is more a human readable input format
MOL - At least it is documented, but why uses fixed width numeric fields?
MOL2 - Documented
PDB - Atom numbers in fixed width fields, number of atoms limited to 99999, creativity a go-go in the atom name field (obviously without putting the element type in the appropriate field)
PDB-Q - A column is 10 bytes shorter, so another reader is needed
POSCAR - Simple format, but the kind of atoms is not in the file
SHEL-X - No big problems, except understanding symmetry definitions
Siesta - No problems
XDATCAR - No problems
XYZ - No problem
XYZ plus unit cell - No problem

The saga continues...

Long time, no post...

2007-01-28T05:47:00.000+01:00

Chemistry visualization made me disappear for a long time...

What happened? Well, the biggest thing is the acceleration for the conversion of my Molecular Visualization Environment STM3 to the new architecture, tentatively (and trivially) called STM4.

STM3 was created as an environment in which new, advanced or unusual visualization techniques could be quickly implemented answering my users requests. But this means that STM3 grows organically. In other words, was starting to be a mess...

The new development goal is mainly to have again a clean architecture in which the new functionalities could be implemented again.

The progresses in this work could be tracked on the STM4 page. My plan is to finish end January...

Free or commercial?

2006-11-02T14:32:00.000+01:00

Last October 17th I was invited to talk at a (rather specialized) chemistry workshop. My talk was about the status of current chemistry visualization tools.

The majority of the audience agreed that was a "provocative talk". And this was exactly my intent.

But a person from a commercial company (producer of various software for chemistry) disagreed. His statement was: "If you pay nothing, you deserve what you are getting". That means usability problems, standard functionalities, etc.

I disagree, also if I find some logic in the argument.

But what do you think? Is it correct to equate free to quick-and-dirty? Is the condition of chem viz software desperate with no evolution possible?

The Perfect Tool Project

2006-10-11T18:47:00.000+02:00

Well, the project has started. Besides the funny name, it wants to stimulate the chemistry researchers that use the CSCS computational resources to think about visualization for their work.

Chemistry is a visual science, but from our experience seems that the chemistry visualization tools lag behind. And seems that the researchers are resigned to this and do not think about new things to be asked to visualization tools. But from another side, seems that when they are convinced of the benefits visualization could provide to their work, they do not stop to ask advanced functionalities for their tools.

This project starts from a survey of existing tools and an analysis of "perfect tools" in other fields. The end could be another tool, but we think the best outcome will be an active dialog between chemistry researchers and visualization experts to advance the whole field of chemistry visualization tools.

If you want more details, go to the Perfect Tool Project page.

The Scientist is a Storyteller

2006-09-29T15:53:00.000+02:00

I always stress the fact that presenting scientific data or describing a scientific achievement needs a good story. The scientist should be also a storyteller that captures the audience attention, that stimulate curiosity and that evoke mental images.

I have just found a short article by Roald Hoffmann (chemist, Nobel laureate and engaging writer) that says the same things, but in a much better way.

The report review agony

2006-09-25T16:39:00.000+02:00

At last i have finished the report "From Tradition to Insight Support - What we can ask to our Chemistry Visualization tools". I'm quite happy of the result. But before this you should pass through the "review agony": the more you read it, the more errors you find. This, plus no one that help and sympathize with you.
Anyway, here are some sparse ideas i want to share with you about the "review agony":

Colleagues are always busy and do not find time to help you reading and correcting your paper. I do not speak about reviewing the content of the paper. This is asking too much. It is simply asking to read and correct the language.
Correcting and revising a paper works better on paper with a red pen. On video is not the same. Why?
It is vital to have a deadline. Everytime you re-read the paper you find things to change or polish. So it is important to force yourself to stop.
A tool like latex2rtf is really useful. Word is really helpful for spotting errors. Well, you have to take cum grano salis its suggestions.
When you find a reference, you should enter it immediately in a Bibtex database. Otherwise entering it during the paper writing is a mess.
Latex is really not so bad. If you do not do anything too complex it is really simple to use.
Remove the "Latex look" from documents produced using Latex. You now, this peculiar font and spacing. that make them immediately recognizable. Well, I added a simple \usepackage{txfonts} and \usepackage{parskip} and everything looks more interesting (try it).
Same source, different result on Linux and Windows. Bleah!

Visualization Tools and Games

2006-09-22T18:23:00.000+02:00

To create more effective visualization tools, we can look to the area of video games. Here the immersion, called "playability", is of supreme importance. The gamers do not want to be distracted by peripheral interaction issues; they want to reach the game goal. They want "fluid interaction" with the tool.

This is exactly the same goal pursued by a researcher looking to an unknown structure. Think about options selection: currently the user should find a menu, find the right menu and push the right option. Instead something similar to a 3D menu widget or semi-transparent display positioned near the attention point distracts a lot less.

I collected some papers and someday I plan to explore the issue (if I can find someone that could work with me)

But...

There are constrains from the operating systems producers that are not interested to change the status quo
There is an uniformity issue: Exit is always under the File menu etc.
If you want the certification sign (Win logo for example) you should follow the interface guidelines of the platform. Maya for example has circular popup menus but no Win Logo.

On the other side:

If you use only one application you do not need uniformity.

Summarizing: it seems the eternal fight between standarization and innovation. But if you want to explore new territories, you need to do something different and look at the results.

From Tradition to Insight Support

2006-09-22T15:32:00.000+02:00

I have just finished writing this report that carries as subtitle: What we can ask to our Chemistry Visualization tools

If Monday, during the CSCS's User Day, it will collects nice comments from our users, it will be somehow published for comment here. For now here is the abstract:

“There has always been a strong relationship between chemists’ understanding of chemical phenomena and the associated external representations.
It is thus common wisdom that visualization tools should play a prominent role in the chemistry discovery cycle, and they do indeed.
But looking at them with a non-chemist eye, reveal that they implement some interesting visualization ideas, but fall short on the perceptual support side and, more important, on the human-computer interaction (HCI) principle implementation.
In this work we substantiate those findings with a tool survey and an inventory of chemistry representation methods. We hope those two results would be the basis for engaging a discussion with the chemist research community to clarify, understand and define the characteristics that an effective chemistry visualization tool should have”.

Starting

2006-09-22T14:46:00.000+02:00

I'm always experimenting with new technologies...

So here is a Blog about Chemistry Visualization. It wants to be as a starting effort to communicate and collaborate about topics of interest of the Chemistry community about visualization tools and methods.

First of all the plan is to publish and share what I'm doing here at the Swiss National Supercomputing Centre (CSCS) with my chemistry users: tools, research, tips, etc. etc.

Beside this, this blog it's an experiment in community building. Will it have success? Time will tell.