Camera for fluorescence detection system

[Charles Maldonado]

Hi all,

I am developing a DIY detection sytem for fluorescence. I want to do the actual detection with a camera, but right now I am struggling a little bit with the selection of this camera There are so many cameras available and I don't know, which one of them will allow me to detect fluorescence. I guess, I need operation in low-light conditions and a high SNR. But aside from that, I don't know, which minimal specs I need...

I just want to use it for detection, not for imaging.

Could you help me please? My max budget is around 500 USD. But the cheaper, the better :D
Thanks!

[S James Parsons Jr]

Hi Charles,

Let's narrow this down, do you have any constraints besides $500 and fluorescence?  Exposure time, weight, ISO sensitivity, filter cost?  A raspberry pi camera comes in 5mp ($8.50), 8mp ($25) and 12.3mp ($50).  How are you illuminating the sample? 

Best,

James

--
-- You received this message because you are subscribed to the Google Groups DIYbio group. To post to this group, send email to diy...@googlegroups.com. To unsubscribe from this group, send email to diybio+un...@googlegroups.com. For more options, visit this group at https://groups.google.com/d/forum/diybio?hl=en
Learn more at www.diybio.org
---
You received this message because you are subscribed to the Google Groups "DIYbio" group.
To unsubscribe from this group and stop receiving emails from it, send an email to diybio+un...@googlegroups.com.
To view this discussion on the web visit https://groups.google.com/d/msgid/diybio/c5a7af8c-03b2-4bfa-af53-8a9e709c5471n%40googlegroups.com.

[Lisa Thalheim]

Hi Charles,

if you only want to do detection, maybe something like the TSL235R would be an option for you - it’s cheap, and it can detect very low levels of light. I’ve used it to quantify fluorescence from DNA dyes like SYBR Green in microliter volumes.

datasheet here: https://www.sparkfun.com/datasheets/Sensors/Imaging/TSL235R-LF.pdf

Kind regards,

Lisa

On 23. Nov 2020, at 19:10, 'Charles Maldonado' via DIYbio <diy...@googlegroups.com> wrote:



--

[Abizar Lakdawalla]

do you have a diagram of your set up. 

BTW, cell phones work very well, you can set up exposure times, etc. 

--

[John Griessen]

On 11/23/20 5:46 AM, 'Charles Maldonado' via DIYbio wrote:
> Hi all,
>
> I am developing a DIY detection sytem for fluorescence. I want to do the actual detection with a camera, but right now I am
> struggling a little bit with the selection of this camera There are so many cameras available and I don't know, which one of them
> will allow me to detect fluorescence. I guess, I need operation in low-light conditions and a high SNR. But aside from that, I
> don't know, which minimal specs I need...
> I just want to use it for detection, not for imaging.

Hi Charles,

I like Lisa's suggestion of a component very much if you are dealing with a zone of light. If you have a cell colony and want to
choose the brightest pixel out of a blob or growing cells, an image sensor (camera) would be good, otherwise if it makes a zone of
uniform light (maybe from a vial or cuvette) it can be lined up with the field of view of a sensor and get good repeatable results
when using a vial to hold "stuff" and get light out of one side of it. What kind of it are you dealing with?

Do you have a way to control where your fluorescing "thing/suspension/slide/dish" is in front of the camera/sensor? For instance
if you have a cell suspension in a particular vial with a certain wall thickness of glass you could calibrate that for a light
intensity and it wold work fine if the shape of it fits somewhere so it always aims a zone of light the same way at a sensor.

--
John Griessen
blog.kitmatic.com Albuquerque NM building lab gear for biologists

[Charles Maldonado]

Thanks a lot for all of your answers. Sorry that I wasn't more specific in my initial post. The 500 $ are only for the camera. Filters are already set up. I have to measure fluorescence of moving particles with at least 20 samples per second (20 fps). Furthermore the sensor should be applicable for a broader spectrum. Right now I only have to detect one fluorophore. Besides that I basically don't have any constraints.
The moving particles are in a tube with fixed wall thickness. These particles are only moving in one direction and have a constant speed. So it's fine that I only measure the fluorescence directly in front of my sensor. I have to detect their presence/absence. I am triggering the fluorescence by illuminating the particles with light from a LED. I want to place my sensor perpendicular to the light source.

[John Griessen]

On 11/23/20 4:23 PM, 'Charles Maldonado' via DIYbio wrote:
> The moving particles are in a tube with fixed wall thickness. These particles are only moving in one direction and have a constant
> speed. So it's fine that I only measure the fluorescence directly in front of my sensor. I have to detect their presence/absence.
> I am triggering the fluorescence by illuminating the particles with light from a LED. I want to place my sensor perpendicular to
> the light source.

Since you say you don't need to know how strongly they glow, or if there are two of them clumped together, you could get away
without image processing.

" I only have to detect one fluorophore. "

Will there be a low low chance of two touching or close to each other?

If so, the sensor route is good. You might even be able to detect when two are in the field of view by relative intensities. Is
the light from one particle varying a lot if it is against the far wall of the tube or near wall? Are you wanting to do any
hardware or code development, or just "full manual" proof of concept?

John

[Abizar Lakdawalla]

almost sounds like capillary electrophoresis - ...

--
-- You received this message because you are subscribed to the Google Groups DIYbio group. To post to this group, send email to diy...@googlegroups.com. To unsubscribe from this group, send email to diybio+un...@googlegroups.com. For more options, visit this group at https://groups.google.com/d/forum/diybio?hl=en
Learn more at www.diybio.org
---
You received this message because you are subscribed to the Google Groups "DIYbio" group.
To unsubscribe from this group and stop receiving emails from it, send an email to diybio+un...@googlegroups.com.

To view this discussion on the web visit https://groups.google.com/d/msgid/diybio/c79b94c9-2518-c6df-5412-4124f6bf4b03%40industromatic.com.

[Charles Maldonado]

" Is the light from one particle varying a lot if it is against the far wall of the tube or near wall?"

Unfortunately, I don't know, whether the light is varying or not depending on the position of the particle. I am developing the system from the scratch and there is almost no experience with fluorescence at my institute. 

"Will there be a low low chance of two touching or close to each other?"

I think, the chance will be quite high...

These particles have the same emission spectrum. But my filter set is interchangeable, so separate measurements with particles with different emission spectra can be conducted.

"Are you wanting to do any hardware or code development, or just "full manual" proof of concept?"

Some code will also be part of my setup. I want to display the sensor data. My initial thought was getting the intensities of my cameras pixels, calculating the average value, normalizing it to some kind of maximum and then plotting it.

Charles

[John Griessen]

On 11/25/20 4:31 AM, 'Charles Maldonado' via DIYbio wrote:
> " Is the light from one particle varying a lot if it is against the far wall of the tube or near wall?"
>
> Unfortunately, I don't know, whether the light is varying or not depending on the position of the particle.

Hi Charles, I asked trying to guess what your goal is -- maybe to separately count particles without counting two as one.

> "Will there be a low low chance of two touching or close to each other?"
>
> I think, the chance will be quite high...

That makes the goal of count-particles-without-counting-two-as-one difficult.

> "Are you wanting to do any hardware or code development, or just "full manual" proof of concept?"
>

> Some code <snip> My initial thought was getting the intensities of my cameras pixels, calculating the average value, normalizing it to some kind of maximum and then plotting it.

But which pixels would mean what if you don't have control over the place they are aiming at, or whether the particles are in that
zone? This seems to be that same goal as is called flow cytometry, which some have charge big bucks for and there are probably
some patents involved if any gear is built for hire. Can you think of anything easier as a goal? It does sound like you want a
microcontroller to read particle sensors and save data to send along when you ask it.

Can the flow rates be very slow and work? What liquid volumes can you pump in a volumetric controlled way?
ONe way to spread the particles out is to make a drawn glass tube thin enough in the middle to let just one or so pass, and keep
the concentration of the suspended cells/particles low enouigh that they don't all rub against each other -- plenty of
interstitial liquid -- so the volume of liquid spreads them out as they go through the narrow drawn part of the tube.

Sounding like what you were thinking yet?

--
John Griessen
industromatic.com Austin TX building lab gear for biologists

[j boogie]

My lab was able to visualize fluorescence (protein of interest) in cells working from some of the instructions found here (Haseloff lab, Cambridge).

https://webcache.googleusercontent.com/search?q=cache:GbEcbAud3GsJ:https://haseloff.plantsci.cam.ac.uk/imaging/cheaposcope/optics/index.html+&cd=11&hl=en&ct=clnk&gl=us