Potato chromosomes 1 to 6, and 7 to 12, with LEDs marking traits of interest |
My contribution was the background images, which are actually drawn using the bases of the potato genome instead of pixels. For this I wrote a little Python script to render photos using A, C, G and T from a FASTA sequence file, using Biopython to load the sequences, the Python Imaging Library (PIL) to load the photos, NumPy to manipulate the array, and ReportLab to render a PDF.
Potato chromosomes 9 and 10 | Close-up showing the A, C, G, T pixels |
Background
Potatoes in Practice is an annual event hosted and co-organised by my employer, the James Hutton Institute (and its predecessor SCRI, the Scottish Crop Research Institute) along with the Potato Council, SRUC (Scotland's Rural College), and others. Billed as the largest potato event in the UK held in a field, there were assorted potato varieties planted on site, large farm machines on display, and lots of farming industry companies and research organisations represented in the marque. Unfortunately I missed the live machinery demonstrations, and there was a disappointing lack of potato themed refreshments this year. The hog roast was tempting but I'd already eaten.
The Potato Genome Display
"The potato genome and how we use it to deliver improved potatoes", The James Hutton Institute exhibit at Potatoes in Practice 2013 |
The red switch for the disease resistance trait LEDs |
- Disease Resistance - red
- Tubers - green
- Abiotic Stress -
yellowwhite - Nutritional Qualities - blue
Materials and Methods
So how did it work? The posters were printed off at A0 size on cloth rather than paper (a very handy trick for getting a poster to an international conference without struggling with a poster tube as luggage), then mounted onto white painted wooden boards (about 5mm thick) with a wrap round margin to staple them securely. Holes were pre-drilled into the boards at the appropriate positions for the LEDs to be mounted, with the wiring hidden at the back. A screw terminal at the bottom of each board was used to allow each poster to be transported individually, and then reconnected - made a little easier by colour coded wires.
Pay no attention to that man behind the This shows the LED wiring, wooden boards and how the cloth posters were wrapped round to mount them |
The small round photos were printed and mounted separately - you can see some peeping through the gaps between the posters in the behind-the-scenes photograph above.
Team
The whole event was a massive team effort. I got involved through a previous collaboration on potato disease resistance genes, or specifically the NB-LRR gene family, where I'd prepared the chromosome diagrams in the paper Jupe et al. (2012) and its follow-up Jupe et al. (in press).
Ingo Hein, Leighton Pritchard and myself came up with the poster chromosome idea and the background image approach in an impromptu brainstorming session - sparked by a related bit of art Leighton had prepared some years before as a journal cover image. Leighton's old code wasn't available, so I rewrote a script to render photos using letters from a genome sequence as pixels. Ingo volunteered to take photos to use as backgrounds (with others coming from the institute's archive), with Leighton advising on what to look for with high contrast edges. Judith on the graphics team prepared and printed the final posters and associated materials. Ian from the workshop team created the LED setup, the switch boxes, and the wooden boards. Ingo also co-ordinated things including gathering the trait information from other collaborators. Then even more people manned the stand, and worked on all the other displays and exhibits.
Really cool script! Thanks for posting it on github. I've modified it a bit for broader usage and reposted it here, crediting you. http://www.haplotypewriter.com/blog/making-an-organism-out-of-a-genome-with-python/
ReplyDeleteAt some point the URL broke, Aaron's blog page is now at http://haplotypewriter.com/2013/08/27/making-an-organism-out-of-a-genome-with-python/
ReplyDelete