It looks like you're new here. If you want to get involved, click one of these buttons!
I'm still assembling the FlyPi and can't give any comments on the functionality yet but I have to get rid of a suggestion of mine
Is there a specific reason why you don't use laser-cutted acrylic parts? If you are looking for a fast method to produce a great amount of FlyPis I would go for a hybrid 3D-printed/laser-cutted model!
Thanks for using the Forums and for your suggestion. We actually have no good reason to be using exclusively 3D printing, other than this is what is readily available for us at the moment.
But you are completely right. As soon as we need to produce more units, this will become very hard to pull off using 3D printers... Problem for that conversion at the moment is the lack of laser cutting know-how and laser cutters available !
We tested the setup with some transformed barley leaves and we stumbled upone some other problems. To excite the GFP in the cells we need the light coming from the same side as the image is recorded (camera and LED have to be at the same side). We only can do that if we put the camera on the top of the FlyPi but then the resolution is not good enough because the camera is too far away.. We will try and use leaves from tobacco to see if we then can excite from the "back" of the sample.
Is there a way implemented to put the LED on the same side as the camera so far?
Interesting, that is something we didn't encounter so far. And now that I think about it, something we should have considered before too... How do traditional systems solve this issue? Dichroic mirrors? I wonder if having the light guided through a fiber all the way to the sample would help the problem...
I forwarded this question to our confocal microscopy expert
I tried the FlyPi with Vektor and first of all I wanted to say that it is really a cool and exciting idea in my opinion. I hope I can give some useful feedback. Most things were already mentioned. As Vektor said, for our purposes we need the UV light on the same side as the camera, since our samples (plant leaves) consist of several cell layers and the UV light can't go trough the sample. For us a upright system with the camera on top works the best. Your assuption was right, usually dichroic mirrors are used to achive optimal results. I am not quite sure, but I think you could also try a set up were the light source comes from a small angle, but you would probably lose some signal in that way.
Beside the set up we came across two bigger problems when we did the testing and basically you already adressed them.
First of all I should mention that we had two differend kinds of samples. The first one was barley leaves with only single cells on the leaves that should express GFP. The second one was tabacco leave discs with an areal expression of a GFP tagged fusion protein. In both cases we unfortunatly couldn't detect any signal. I think the reason could be that the UV light is to weak and maybe a concentration of the UV through fiber optic cable would help.
Also for the transformed single cells the resolution wasn't good enough. I don't know if that can be fixed by bringing the camera closer to the sample or if you need some additional lenses (probably). In case of barley cells they are about 200µm long.
In the end I have one minor last point. It would be nice if you could make it so that you can change the width of the clamp which holds the objective slide. I cut my finger trying to bring the objective slide into the clamp
I hope this was helpful.
Thanks a lot for the feedback. It is useful
Sorry to hear about your finger. Hope it wasn't anything bad!! Also it is good to know about it, since it is something we need to consider if we succeed in getting this into more hands out there (specially if it includes people in schools and the general public, who have less experience with microscopes and microscope slides..)
I'll be visiting the institute soon for an event Vektor is putting together, and if you have time and don't mind, would it be possible for me to pop by and take a look at the samples you are imaging? (first out of sheer curiosity, second because I suspect the led you guys got could be better for GFP - Is the light blueish or more to purple?), we've also switched the camera in the system which could also improve things - I'm a bit intrigued about not being able to image 200µm long structures... In principle it should work. I know this will sound like a silly question, but did you twisted the camera lens further out? it should increase magnification quite a bit.
don't worry, it was just a little cut. Regarding the resolution, maybe I was a little imprecise. It should be ansolutely possible to see the cells, but you would probably just see some glowing dots. To see structures on or in the cells the magnification is not sufficient. But maybe I will try around on thursday again a little, because although we twisted the lens, we just put the sample into focus. I will check if I can change something there. The led we have shines blueish.
I will also join the event on 27th and of course we can have a look at the samples then. I will prepare some barley samples, but we will probably have no tobacco.
Looking forward to meet you there.
Something small but worth changing is on the material list for the arduino socket terminal strip you have it listed as 20pin, linked to 10 pin and in actual fact you need 15pin so it fits the PCB board and arduino
thanks for spotting this! Could you please tell us which material list you used? (I mean from which repository you took it?)
From the github material list:
cool thanks. I can change that. If I remember correctly we couldn't find a 15 pin one from that supplier, and we wanted to stick to one supplier so that it would be easier for people in Europe/Germany to get all components.
In the meantime we integrate a 1-click BOM option (leveraging the great work from kitspace: https://kitspace.org/boards/github.com/prometheus-science/flypi/