Difference between revisions of "Assignment 2: fluorescence microscopy"

From Course Wiki
Jump to: navigation, search
 
(100 intermediate revisions by 3 users not shown)
Line 3: Line 3:
 
[[Category:Optical Microscopy Lab]]
 
[[Category:Optical Microscopy Lab]]
 
{{Template:20.309}}
 
{{Template:20.309}}
−
[[Image:Galileo Galelei.jpg|center|300px]]
 
  
−
<blockquote>
+
This assignment has 3 parts.
−
<div>
+
* [[Assignment 2 Part 1: Noise in images|Part 1: Noise in images]]
−
''It must be explained how it happens that the light is conceived into the stone, and is given back after some time, as in childbirth''
+
* [[Assignment 2 Part 2: Fluorescence microscopy|Part 2: Fluorescence microscopy]]
 +
* [[Assignment 2 Part 3: Build an epi-illuminator for your microscope|Part 3: Build an epi-illuminator for your microscope]]
  
−
<blockquote>
+
Submit your work on Stellar in a single PDF file with the naming convention <Lastname><Firstname>Assignment2.pdf.
−
''&mdash; Galileo Galilei on the unusual properties of the stone ''lapis solaris''. ''
+
−
</blockquote>
+
−
</div>
+
−
</blockquote>
+
−
<br/>
+
  
 +
{{Template:Assignment Turn In|message= Here is a comprehensive list of what you need to turn in:
 +
Part 1
 +
# Turn in your plot of simulation results
 +
#* Find the interval [ <math>\mu - s</math>, <math>\mu + s</math> ] that contains about 68% of the simulation results.
 +
#* Turn in the code you used to find the interval.
 +
# On the same plot, plot the PDF of a normal distribution with the same mean and standard deviation as the simulation results. Multiply the PDF by a constant so that it has the same vertical scale as the histogram.
 +
# Is the Poisson distribution a good approximation of shot noise?
 +
#* What percentage of the results fall with one, two, and three standard deviations?
 +
#* Produce a log-log plot of standard deviation of the number of photons emitted as a function of the average number of photons emitted for probability of photon emission equal to the values: 10<sup>-6</sup>, 10<sup>-5</sup>, 10<sup>-4</sup>, 10<sup>-3</sup>, and 10<sup>-2</sup>.
 +
#* Hints: Use a nested loop. Use the poissrnd function instead of the line containing rand. Speed things up by getting rid of the plotting inside the loop and only run 100 simulations for each probability.
 +
#* What is the relationship between the number of photons detected and the noise (standard deviation)?
 +
#* Turn in the code you used to generate the plots
 +
# On one set of axes, plot the variance vs. mean for exposure times of 10<sup>-4</sup>, 10<sup>-3</sup>, 10<sup>-2</sup>, 10<sup>-1</sup>, and 10<sup>0</sup> seconds.
 +
# Using the camera measurements provided:
 +
## Plot the raw data from the static scene measurements and the model best fit on one set of axes.
 +
## Calibrate the gain setting: make a plot of the actual camera gain in electrons per ADU versus the software setting.
 +
## Provide a formula for converting the camera gain setting to the actual gain value.
 +
## Plot dark current versus exposure time and determine the value of ID in units of electrons per pixel per second.
 +
## Determine the read noise standard deviation.
 +
## Under what circumstances is each of the three noise terms is dominant?
  
−
==Overview==
+
Parts 2 & 3
−
Fluorescence microscopy is one of the  
+
# Draw a block diagram of the LED epi-illumination path. Indicate the focal lengths of all lenses, the correct lens orientation, and all important distances between components.
 +
# Lenses L3 and L4 make an image of the LED. Assuming the initial size of the LED is 1.3 mm, what is the size of the LED image made by lens L3?
 +
# For each bead sample, include the original, reference, and flat-field corrected images in your lab report. In the caption note the exposure and gain settings used for each image.
 +
# For one set of images (either the 0.84 or 3.6 &mu;m beads and their corresponding dark and reference images), include the MATLAB code you used to calculate the flat-field correction.
 +
}}
  
−
 
+
{{Template:Assignment 2 navigation}}
−
[[Image:Alexa 568 and filters.png|thumb|260px|right|Excitation and emission spectra of Alexa 563. Green stripe indicates excitation laser wavelength of 532 nm. Shaded area shows passband of emission filter.]]
+
−
In this part of the lab, you will add the hardware necessary to make epi-illuminated fluorescent images with your microscope and make a few test images of plastic, fluorescent beads. A typical spectrum is shown on the right. The excitation light source is a 5 mW diode laser with a nominal wavelength of 532 nm &mdash; a striking, brilliant green. Emission is in the red/orange range. To make the correction for nonuniform illumination, you will also make images of a uniform fluorescence reference slide and a dark image with the illuminator turned off.
+
−
 
+
−
[[Image:20.309 130911 YourMicroscope.png|thumb|260px|right|Fluorescence microscope block diagram.]]
+
−
As shown in the block diagram, the major components required for fluorescence imaging are an illuminator (laser, L3, L4, and L5), dichroic mirror (DM), and emission filter (BF). The illuminator provides light in the appropriate wavelength range to excite the fluorescent molecules in the sample. Fluorescence microscopes that use broadand light sources such as arc lamps require an additional filter called an excitation filter to limit the wavelengths in the illumination to the proper range. Because lasers emit light in a very narrow range of wavelengths, an excitation filter is unnecessary.
+
−
 
+
−
[[Image:20.309Hw3Imagespectra.JPG|right|thumb|260px|Transmission spectra for the 565DCXT dichroic and the E590LPv2 barrier filter]]
+
−
Excitation light comes from beneath the sample, through the objective lens. A dichroic mirror directs the excitation toward the objective and sample. The mirror must reflect wavelengths in the excitation range and pass the longer wavelengths of the emitted fluorescence. In fluorescence imaging, illumination intensity is typically 5 or 6 orders of magnitude greater than emitted fluorescence, so it is crucial to filter out excitation photons as completely as possible. The dichroic mirror passes a substantial amount of green light, on the order of five percent. The barrier filter does a much better job of removing the green light, attenuating the excitation wavelengths by about 5 orders of magnitude. The barrier filter is essential for making crisp, high-contrast images.
+
−
 
+
−
To provide collimated illumination in the sample plane, light from the illuminator is focused at the back focal point of the objective.
+
−
==Assignment details==
+
−
 
+
−
This assignment has 2 parts:
+
−
 
+
−
# [[Assignment 2: epi illuminator for fluorescence microscopy|design and build an epi illuminator for your microscope]]
+
−
# [[Assignment 2: resolution and noise|develop a software simulation to explore the concepts of resolution and noise]]
+
  
 
{{Template:20.309 bottom}}
 
{{Template:20.309 bottom}}

Latest revision as of 18:23, 14 February 2020

20.309: Biological Instrumentation and Measurement

ImageBar 774.jpg


This assignment has 3 parts.

Submit your work on Stellar in a single PDF file with the naming convention <Lastname><Firstname>Assignment2.pdf.


Pencil.png

Here is a comprehensive list of what you need to turn in: Part 1

  1. Turn in your plot of simulation results
    • Find the interval [ $ \mu - s $, $ \mu + s $ ] that contains about 68% of the simulation results.
    • Turn in the code you used to find the interval.
  2. On the same plot, plot the PDF of a normal distribution with the same mean and standard deviation as the simulation results. Multiply the PDF by a constant so that it has the same vertical scale as the histogram.
  3. Is the Poisson distribution a good approximation of shot noise?
    • What percentage of the results fall with one, two, and three standard deviations?
    • Produce a log-log plot of standard deviation of the number of photons emitted as a function of the average number of photons emitted for probability of photon emission equal to the values: 10-6, 10-5, 10-4, 10-3, and 10-2.
    • Hints: Use a nested loop. Use the poissrnd function instead of the line containing rand. Speed things up by getting rid of the plotting inside the loop and only run 100 simulations for each probability.
    • What is the relationship between the number of photons detected and the noise (standard deviation)?
    • Turn in the code you used to generate the plots
  4. On one set of axes, plot the variance vs. mean for exposure times of 10-4, 10-3, 10-2, 10-1, and 100 seconds.
  5. Using the camera measurements provided:
    1. Plot the raw data from the static scene measurements and the model best fit on one set of axes.
    2. Calibrate the gain setting: make a plot of the actual camera gain in electrons per ADU versus the software setting.
    3. Provide a formula for converting the camera gain setting to the actual gain value.
    4. Plot dark current versus exposure time and determine the value of ID in units of electrons per pixel per second.
    5. Determine the read noise standard deviation.
    6. Under what circumstances is each of the three noise terms is dominant?

Parts 2 & 3

  1. Draw a block diagram of the LED epi-illumination path. Indicate the focal lengths of all lenses, the correct lens orientation, and all important distances between components.
  2. Lenses L3 and L4 make an image of the LED. Assuming the initial size of the LED is 1.3 mm, what is the size of the LED image made by lens L3?
  3. For each bead sample, include the original, reference, and flat-field corrected images in your lab report. In the caption note the exposure and gain settings used for each image.
  4. For one set of images (either the 0.84 or 3.6 μm beads and their corresponding dark and reference images), include the MATLAB code you used to calculate the flat-field correction.


Back to 20.309 Main Page

Retrieved from "http://measurebiology.org/w/index.php?title=Assignment_2:_fluorescence_microscopy&oldid=43061"