Deep Data Prototype 2: arabidopsis

From the beginning to the end in one post….

A few months ago I was invited to make a new work for an exhibition called “Sin Origen Sin Semilla” which would take place at UNAM in Mexico City. I decided that this new work would be the long overdue second piece in the Deep Data series, and this work would use the model organism plant Arabidopsis.

Consistent with the Deep Data concept, this work would reproduce or reinterpret space bioscience experiments using the same organism. It appears that arabidopsis has been used extensively for gravitaxis and root development experiments, and despite my future plans to build a random positioning machine, that was not feasible on the short timescale I had to develop and make the piece.

Being a model organism, arabidopsis has numerous documented naturally occurring and genetically manipulated mutants. One of these is the photomorphogenic mutant NW67, which will show variations in the development of the hypocotyl depending on the spectrum of light it receives at the seedling stage.

To cut a long story short, Deep Data Prototype 2 became a piece in which arabidopsis seedlings could be grown under the light conditions, both real and imagined from science fiction, on other planets and moons in the solar system – planets and moons where there has been some discussion of the viability of human bases and/or alien life.

The growth systems for the seeds/seedlings would be heavily indebted to the RootChip developed by Guido Grossman et al, a PDMS device which would allow nutrients to reach the developing root system while allowing full observation of the growth process.

The original 3D visualisation of how the piece would look.

3D_2

From drawings made in Inkscape and Draft Sight all the wooden pieces for the arms and growing boxes were cut on the laser cutter in the Laboral Fablab.

DSC06465

DSC06467

then assembled and sprayed a deep matt grey

DSC06497

To make the PDMS growing devices a positive was made using the CnC router on a block of wood. From this positive a silicon mold was made.

DSC06475

It looked great, but for some reason I’ve yet to discover the PDMS would never set. 4 months later it is still tacky!

So eventually I resorted to the tried and tested petri dish casting vessel. I made a positive of the 4 pronged fork shape and stuck it to the bottom of a petri and cast it 8 times. it didn’t come out so beautifully but at least it worked. I then cast 8 2mm thick sheets to blank of the bases and seal the microfluidic system.

DSC06521

DSC06522

Cut off pipette tips were set into the PDMS wells, placed within a retaining ‘ring’ and connected to a tube from the nutrient reservoir.

DSC06518

Lighting gel filters which accurately match the known light conditions on Earth, the Moon, Mars, Venus and Titan, as well as how they had been envisaged in popular science fiction art, were placed into the holders on the top of each growing box, and a full spectrum (5,500K) LED bulb suspended above.

DSC06519

The bulbs were connected 4 each to two 12V LED bulb ballasts.

A business card sized magnifying sheet was place in the front of each growing window and the nutrient reservoir to aid viewing the tiny seedlings as they grew.

Finished studio view

DSC06511

Finished exhibition view

_MG_3106_7_8Adjust

_MG_3103_4_5Adjust

Finished Temperature Gradient

I’ve been so busy making stuff recently that I haven’t had time to blog it as I go along. Meaning that all the stages between early tests for this device and the finished article are pretty much undocumented. This is a summary of what happened….

 
The large fan and heatsink combo worked great, and this seems to be the secret to successful use of peltier devices – get rid of excess heat as much as you possibly can. I was running the thing at 7.5V, which used about 3.75A and made the metal cold enough to cause condensation on it. The transmitted cold made the end of the fly runs about 12 to 14 Celsius, or a good 10 degrees below ambient, and a good 6 to 8 degrees below optimal drosophila temperature. Good enough.

 
Heating the other end was achieved through a 470 ohm 3 watt resistor which brought the end of the metal runs up to about 28 C. So my range was about 28C to 14C with somewhere around 21C in the middle where the flies go in. The thermometers were still problematic, in that they combined the ambient temperature with the metal temperature so these figures my be 3 or 4 degrees out. It turns out the measuring surface temperatures without an IR thermometer is a bit tricky. The most important bit was that I knew that the cold end was a few degrees below the normal preferred range for drosophila, and that was enough for the experiment to function.

 
A PS3 Eye camera was mounted above the device, and a 10mm clear acrylic top put machined and fixed on so that the flies had to stay in the runs. The camera would be used for filming the fly behaviour during the experiment and for future video tracking and analysis of their movement and temperature zone preferences. Access points were made to enable placing of the flies in the runs, and these were just 6mm holes with neoprene bungs. An orange line was drawn across the top to demark the cold limit. Any flies spending a ‘reasonable amount of time’ would be deemed as cold preferring and selected for breeding.

 
Everything was wired up – one 7.5V power supply for the peltier, and a 12V for the fans and heating element.

 

 

 

 

next up…. experimenting with flies…

Deep Data Prototype_1 base cover

The loose aesthetic inspiration for this new version of the Deep Data Prototype_1 is the Biopan module which is used for living organism experiments in space. Biopan 6 was the mission on which the TARDIS experiment took place to investigate the reaction of tardigrades to the ‘raw’ space environment.

 

 

The original plan was to make a metal cover for the wooden base which would reflect many of the design and mechanical elements of the Biopan module. Unfortunately due to lack of appropriate machine tools, and lack of time waiting for new ones, I decided to use 2mm white acrylic sheet and simplify the design. This is the PDF file from which the code for the laser cutter was used to fabricate the piece.

 

Rebuilding Deep Data Prototype_1

The chameleon nature of this piece keeps going….
For the upcoming exhibition Soft Control exhibition in Maribor this November the piece is being reinvented yet again. For this I’m building a new base which is loosely borrowing from the Biopan module that tardigrades flew on in 2004. Making full use of the FabLab downstairs the wooden base is milled on the CNC routing machine, and will have a matching steel plate on top. The components for the microscope and device stand are being made on the 3D printer.

the wooden base routed from 16mm MDF. cut outs for mac mini, microscope and device stand, and power supply.

 


blender rendering of stand base, exported as .stl file for makerbot.

 

and the printed object in fetching blue ABS

 

mac mini, power supply and stand base located in their positions.

Temperature Gradient

A device for selecting Drosophila that have a preference for cold….
This is the apparatus that will be used for temperature conditioning and selection. Its basically a slab of aluminium with routed covered tracks in which the flies are placed. One end of the device is cooled and the other is heated, aiming for the flies’ ideal temperature zone in the middle. Thus it can be easily noted which individuals, if any, gravitate towards the hotter or cooler end. In my case, any flies gravitating towards the cooler end will be selected for future breeding.

This is the slab after being routed on the computer controlled CNC in the fablab. One temperature probe is measuring the ambient temperature of one end of the slab.

 


The cooling will be provided by a 12v peltier element attached to one end. Needing effective cooling means that the hot side needs to be well set up so as to get rid of excess heat. The delta temperature of the device is around 65 degrees, so I’ll need to get the hot end down to 75 to get my target 10 at the slab. This little heatsink didn’t do the job at all.


nor did this bigger one with a fan on top…

As the hot side heats up, it is enable to dissipate any heat, meaning it just gets hotter and hotter, meaning that heat then begins to transfer to the cold side, meaning that it warms up, meaning that eventually the whole thing gets so hot that it just packs in.

So i’m now trying a big 10cm square heatsink with a 12cm fan sat above it. I’m getting a constant very cold side on the aluminium mount, so its looking good. However, the temperature probe seems to be affected by the surrounding air temperature and gives me a reading of about 17.5 degrees when its obvious that it is much, much colder.
Still some work to do, but its getting there…