Single Gene Gives Mice Three-Color Vision

maynard writes "A study in the peer-reviewed journal Science shows that mice transgenetically altered with a single human gene are then able to see in full tri-color vision. Mice without this alteration are normally colorblind. The scientists speculate that mammalian brains even from animals that have never evolved color vision are flexible enough to interpret new color-sense information with just the simple addition of new photoreceptors. Such a result is also indicated by a dominant X chromosome mutation that allows for quad-color vision in some women." A sidebar in the article includes a nice illustration of what two-color vs. three-color mice might perceive.

Question I couldn't get from the article

[reezle]

Did they provide gene therapy to the mice which then gained color vision, or did they alter the mice before birth? Is it possible to insert genes into an adult organism and permanently change their DNA structure?

True colour

[LiquidCoooled]

I had my first taste of this recently.

We live in a colour society and think when we point a camera at a target and click we take a faithful picture of it.
I was wrong.

I have pictures taken from a recent concert where the camera saw one colour (blue) but the actual colour was violet, it was strange holding it up and seeing it filtered then moving it out of the way to see the real colour.

does anyone know if there are such limitations with original developed camera film, or is it just not noticed?

possible use in humans?

[dunkelfalke]

is it possible to genetically alter humans to make them tetrachromats, thus making them able to see UV like fishes and birds do?

Martian colours

[Dogtanian]

One issue I find interesting in this context is the guy who was colour-blind (that is, he couldn't differentiate colours in certain parts of the spectrum). This guy had synesthaesia [wikipedia.org], and although he couldn't physically see certain colours, he could experience them through his synesthaesia. He referred to them as "Martian colours".

The interesting implication here is that the GM mice's brains apparently developed with the ability to process the new colours. It would be fair to assume that ordinary mice's brains did not even contain the "concept" or "perception" of red hardwired in, since what would the point be?

Thus, if the converse is true, and human brains develop the same way as mice's, it could be assumed that the brains of people with the *physical* inability to detect certain colours from birth would never develop the mental concept/sensation of those colours. (*) But then, now does this explain "Martian colours"?

(*) (If you're having trouble understanding what I mean, try to imagine what ultraviolet "looks" like. Darklight (UV lamp) special effects don't count; that's *visible* light produced when UV hits special fluorescing material. And you can't "cheat" by imagining in terms of false colours (since that, by definition, is *converting* UV to visible-range colours). No, I want you to try to imagine what colour actual UV light would look like... and you'll fail because you've never directly seen UV light, and the concept isn't wired into your brain).

Re:True colour

[Jott42]

Most definitly. No color reproduction technology in existence has the capability of reproducing all the colours that the human can experience. (i.e. the Gamut, nice reading in Wikipedia: http://en.wikipedia.org/wiki/Gamut [wikipedia.org] )

Still no proof of 'full' spectrum vision

[nietsch]

Although their GM mouse made M and L type cones in their retinas, it is still not clear if what they reacted to was only a change in intensity, or if they could see a true difference between the two colors. Normal mouse are essentially colorblind in that region of the spectrum, red triggers the M receptor, but not very much, so you need a brighter red light to stimulate the M receptor equally as greenish light. Since you need a very good control, the test setup was such that normal could not see a difference between the red and green light. Their GM mouse were much more sensitive to red, so to them the red light must have had a much brighter intensity. But that does not mean their brains had adapted themselves to differentiating between red and green light. To test that you would have to measure the sensitivity of the new red receptor and adjust your intensity to that so that the only difference is in the color, not the intensity. The problem offcourse is that you cannot do that same experiment with normal mouse which have a different red sensitivity, and no control == bad science.
So their claim that the GMs mouses brain really processed the red light signal different from the green might be a bit over the top.

(hmm thinking about it, if the GM mouse cannot discern between red and green, there might be a certain redlight intensity where their scores would drop significantly, while the controls would score better. If you cannot find that, my hypothesis is wrong and their claim is right. Now lets see if I can find if they did that test...)

Re:True colour

[19thNervousBreakdown]

Violet is especially tricky. Its wavelength is shorter than blue, but in addition to stimulating your blue cones, your red cones are also slightly sensitive to it. The camera, however, sees the pure, very deep blue. Then, when it goes to display it on the LCD, it only turns on the blue pixel instead of the blue and a little red.

Another thing that people don't generally notice is that the RGB pixels or phosphors don't match up perfectly with everyone's cones. The only way I can think of to have faithful color representation is to have one "pixel" on both camera and display that is sensitive to and can emit any visible frequency of light, with perfectly flat response. IOW, maybe flying AI-controlled cars will have a camera/display like that.

Colorblind posters wanted

[Scrameustache]

What do those graphics look like to you?

Re:Martian colours

[toonerh]

The occipital lobe of the brain (visual processing) - even in adults - can retrain itself to flip the view after wearing inverting glasses, ignore the distortion from "progressive" glasses (for old people like me) and quickly compensate for different colored lighting.

It seems quite possible a mouse's brain could classify groups of cones, especially since they would be obvious from birth on.

Re:Question I couldn't get from the article

[Tatarize]

What you really want to know is if they develop some of these genes to give you superpowers can you have them or do we need to genetically engineer ungrateful children to be able to whoop us.

The latter is the case. Your eyes are destined to suck forever. You can't see infrared or ultraviolet, you can't see like a hawk, nor can you get the lungs of a bird, the electro-sensing power of a platypus, ability to freeze solid like a toad, smell things as well as a dog, hold your breath like a whale. Even simply fixes like giving humans the ability to make their own vitamin C (every mammal has that save great apes and guinea pigs). No fixing the mammal eye so the all the blood and nerve don't run in front of the lens. No fixing the recurrent laryngeal nerve so that it goes from the brain straight to the larynx rather than looping around the aortic arch for no reason at all.

We could however, perhaps give such changes to our kids, those ungrateful little snot-filled twerps. You'll have to live being a social thin-haired ape who can play with fire and kill just about anything after making the tool for the job.

The Ducks Win It!

[DynaSoar]

Ducks are pentachromats. They have 5 different receptors for color. That doesn't mean they see other colors than we do, but it does mean they have better color differentiation. I can think of no other explanation other than ducks evolved from artists.

Maybe we can put them to work testing monitors. Your garden variety graphics card and monitor are already capable of producing more colors (4.28 million or some such) than humans can differentiate (3 to 3.5 million).