Hmm, I see difficulty in that, without data and the experiment done, there is high risk that bringing in the artists, visualization experts, etc. could be a big waste of their time. *If* you could know a big study was going to be worth it, this is the best approach. In reality there exists a high chance a study will fail to meet expectations, and making this approach too risky; I think it would be seen as putting the cart before the horse, so to speak. - Chris Lasher

I have mixed feelings on that. All for being inclusive, but it needs to have some substance - Deepak Singh

Absolutely agree that people with these skills are valuable, but agree with Chris; I don't see how you can decide on analysis/visualisation until data are collected and scored. - Neil Saunders

Indeed these people are valuable, but I think it is important to bring in the right people at the right time. If I don't yet know what my result is, what would a graphics artist be drawing? What would a communicator communicate? I do agree that having visualization scientists involved early makes sense, though. - Lars Juhl Jensen

I used one for my thesis. Thought doing 3d graphics representations (done mostly in FormZ) made more sense than the stale figures most used. Think it was successful. But this was after all data were collected and analyzed - Deepak Singh

A separate point, statisticians are notably missing from the list. I've been told by statisticians that the chronic mistake made in the biological community is that we come to them after we do our experiments, rather than consult them during the experiment design process to ensure the experiment will have enough samples, timepoints, replicates, etc. from which to draw statistically valid conclusions. The end result of an experiment is much less tractable; sometimes a lost cause. - Chris Lasher

Chris nails it again. Numerous times people come to me with their data, asking about statistics. Almost always, they've designed the experiment in a way that precludes meaningful statistical analysis and it's too late to go back and do it properly. Really bugs me when people view stats as a miracle cure for bad design, then blame me when I refuse to touch their results. - Neil Saunders

It works both ways. I know of one anecdote where a statistician was given some differential expression data and brought it back with everything upregulated (he had assumed up and down were the same and only magnitude mattered) - Deepak Singh

And where can I find an artist ? (who would understand my data and of course would work for free ). But I like the idea. I remember, some years ago in my previous lab , there was a woman whose job was to make the illustration for the articles or the slides (nice job). But soon everybody installed a software for drawing on their PC and made their graphics by themselves... - Pierre Lindenbaum

Even with my limited knowledge of statistics I still remember one of the first words I heard on the "statistics for biologists" lecture: in biology, infinity starts at 30 :) Anyway,the quote resonates with me, although in a different way. I think the best we could do is to get all these people to make a platform for understanding the experiments. And this should be done not every time in every group but once and released openly ;) - Pawel Szczesny

I do start quite often a small work on somebody's else problem. If such person is doing some stuff for the two years, I have no way to immerse into the project within less than few weeks. Having the project presented in some meaningful way (visualization), that I can understand (communication) and is graphically attractive (graphics) would be of enourmous help not only for me, but to the whole group (having big picture). - Pawel Szczesny

I wonder if an effort like this could be generalizable. Could we reinvent the way microarray data, gwas, etc is displayed? Or does every lab need an artist-in-residence? Can scientists become more literate in visual communication? Would that be enough? - Sutee Dee

I would argue that the peopple who are developing tools (who of course need to be properly supported) should have people with graphic design/UI skills. That seems to me the right place to have these people. There have been some big contributions as I understand it to the visualisation of MRI etc. imagiing by anatomical artists. But I don't think except on very specific projects there would be enough to justify a full timer on each one - Cameron Neylon

I think we, as scientists, tend to systematically undervalue the role of professional communicators - visualization, typography, graphic design, illustration - throughout our work. Many, or most, of our data analysis techniques are more-or-less inseparable from visualization - curve-fitting? Dimensionality reduction? So I'd argue that on the whole visualization isn't about being pretty, it's about transforming a problem into terms we understand, and that's a kissing cousin of theorising... - Andrew Walkingshaw

Read the books of Edward Tufte, or attend a presentation: http://www.edwardtufte.com/tufte... . It won't make you much smarter, but it might tell you how dumb we are about the importance of design and visualization. His bottom-line: Most of us should just copy from the best. - Brian Haugen

I had the opportunity to attend a Tufte seminar when he was in Seattle. I left feeling empowered to change the (visualization) world, but there are still considerable challenges in translating his ideas to the specific needs of biology. For those familiar with Tufte, which of his ideas do you think are the most applicable to biological visualizations? - Sutee Dee