22 July, 2008

The final verdict ...

In a previous post I promised to do some more genome browser screenshot counting. So, that is what I did last week at the XX International Congress of Genetics 2008 in Berlin. I limited myself to the second poster session of the conference that should have contained 675 posters. To my surprise a vast amount was missing though, so I estimate that the number I looked at was closer to somewhere between 400 and 500. Compared to the Barcelona conference the result was poor; I identified only 4 posters with Ensembl screenshots as well as 4 posters with UCSC Genome Browser screenshots and none with NCBI Map Viewer screenshots. So, based on the combined results from two genetics conferences, it seems that the Ensembl and UCSC browsers are about equally popular amongst poster-making geneticists.

However, I had expected more genome browser screenshots in general. What can be the reason for these low numbers? Is there no need for screenshots at all? Or can people not get what they need for their poster from Ensembl or UCSC? We are curious about your thoughts and views on this and are welcoming any suggestions for improvements to Ensembl that will make preparing figures for your poster (or publication) more of a breeze!

4 comments:

Optimistix said...

Come to think of it, counting screenshots will likely lead to a lot of "false negatives" - given that space is at such a premium in posters (and sometimes even in papers), people may not use screenshots unless they really need to. For instance, we used the UCSC genome browser for a paper, and cited it.

How about counting citations to the relevant papers (maybe only in the past year or two)? A quick search using Google Scholar suggests that both are doing exceedingly well, as one would expect.

Bert (Helpdesk team) said...

Hello Rileen,

I completely agree that counting citations (and looking at web usage statistics) indeed is a much better way to get an idea about Ensembl usage than looking at a few hundred posters at a conference. In fact, this is something we already are doing.

Remains the fact that we are open for any suggestions that will help those people who do want to include Ensembl screenshots in their poster or paper.

Greetings from Hinxton

Optimistix said...

Hi Bert,
I hope to be able to take you up on that someday :-)

Anonymous said...

Hey bert,

I was working with a 21-exon (22 if you count some UTR splicing) gene for a while. As it turned out, using the Ensembl screenshot was a bit of a hinderance in many cases because

1.) our gene was on the (-) strand, so it was unintuitive to present useful information addended to the Ensemble screenshot

2.) there was no good intermediate zoom level that would show enough intron/exon boundary detail for our purposes while still being small enough to see the entire gene length. If I zoomed out far enough to see the entire gene, the exons were pixelated together and introns could not be seen.

So I've manually gone through and created a couple of amalgamations of several different screenshots of the Ensembl browser which are suitable for some of my powerpoint presentations.

My advisor has often told me to just hand-draw (or hand-Illustrator) appropriate figures rather than use the Ensembl screenshots, perhaps because the genome browser is pyschologically (at least for us) an early-stage raw-data tool rather than a polished, show-my-data tool.

Just a thought.