BME100 f2014:Group33 L6

From OpenWetWare
Jump to navigationJump to search
BME 100 Fall 2014 Home
Lab Write-Up 1 | Lab Write-Up 2 | Lab Write-Up 3
Lab Write-Up 4 | Lab Write-Up 5 | Lab Write-Up 6
Course Logistics For Instructors
Wiki Editing Help


Name: Emigdio Ruiz Esquivel
Name: John Cunningham
Name: Wandasun Sihanath
Name: Chris Adams
Name: Tim Snelling
Name: Breanna Falcon


Bayesian Statistics

Overview of the Original Diagnosis System There were 34 groups in the BME 100 lab each testing 2 different patients, giving a total of 68 patients. Every team got DNA samples of two different patients to test for SNP. Every group tested about three controlled samples of both negative and positive. In order to analyze them, there was a PCR machine that did this and then used the ImageJ system that would give us a good analysis of the drops. This image program gave us a good reading for each sample and minimized the error with its measures. After analyzing everything with each group, we had to input the data on to a spread sheet that would allow all of the classroom to make a final analysis. This made it possible to ensure an accurate analysis, since each patients DNA was tested at least three times. This also made it easier to make sure the whole lab class was concurring with each other. As long as the trials were similar to each other the data was accurate concerning whether the patients were positive or negative. Since so many trials were done by multiple groups; the amount of human error was decreased along with any errors that could have resulted from the Open PCR machine or the ImageJ software. Three images were taken of each sample to ensure that accurate data could still be gathered even if one or two of the other pictures seemed blurry. Everyone had the same procedure for the set up of the Open PCR machine as well as a positive and negative control. For our group we had the same people who had taken the pictures rename the pictures according to the patient type and trial, this way we were sure that the pictures accurately reflected the data and also renamed them to reduce confusion. The final data that was gathered was as accurate as it could have been because it functioned like how a real lab would in real time. Sometimes results are inconclusive and there were also blank spaces where some data was not recorded. Everyone in the lab tried to reduce the amount of errors preformed in the lab and so the data that was taken from other teams was given the benefit of the doubt that there results were as stated.

What Bayes Statistics Imply about This Diagnostic Approach Both calculations 1 and 2 gave values close to the 1.00 (100%) value that would be ideal, which implies that the diagnostic test itself is relatively reliable. There were a few possible errors however which could have changed this value, some of which could have been that there were wrong mixtures, since some groups had values that were inconclusive, which would have been a human error. Another error could have occurred from slight movement of the camera since it wasn't in a fixed position in respect to the device during the testing. One last error could have occurred from unwanted light exposure from the camera, depending on the type of phone used, which could change the observed results from the tests. For a possible detection step that affected the Bayesian Statistic values, if the camera took blurry or fuzzy pictures, the results may not have been differentiated from one another thus creating faulty data and faulty statistics. As for calculations 3 and 4, the values were very small and not close to 1.00 (100%) which implies that as a diagnostic, the test is inconclusive and doesn't support whether or not someone would or wouldn't develop the disease given a positive or negative result, respectively.

Computer-Aided Design

TinkerCAD is an online application that is similar to SolidWorks because they can both amount to a finished product that can be sent to a 3D printer once they are finalized. TinkerCAD is already pre-loaded with models and different geometric shapes that can be used to build different designs. A design made from scratch or a pre-existing one that is added onto can be made. The website offers free tutorials to learn how a certain design can be created, even if the exact shapes are not present. All that is required to make a design is a shape that is similar to it and from there it can be manipulated into the desired product. The images of the Open PCR machine could be imported into TinkerCAD and served as a template for the changes that we wanted to make for the modified version of the open PCR machine. The finished product looked realistic since TinkerCAD is a 3D software that employs the use of holes and a ruler that can help measure the dimensions of the product to make sure it comes out connected and that there are no gaps featured within the product. For our design the program SolidWorks was used instead of TinkerCAD to demonstrate the changes that we would make to the fluorimeter.

Our Design

The design featured incorporates a cradle that is actually attached to the fluorimeter in order to decrease the inconsistency in measurements. This way if the fluroimeter is accidentally nudged than the whole device will move as one. By attaching the cradle to the fluorimeter, the measurements will be more concise and easy to spot. This gets rid of the annoyance of having to continually measure the distance the light box and cradle were moved and also helps the images appear more clear and not blurry. It is important that the pictures come out more clearly, since they are needed for processing in ImageJ. The fluorimeter will still be able to be adjusted to ensure that the blue LED light on the fluorimeter is able to focus on the drops that are placed on each of the slides.

Feature 1: Consumables Kit

Weaknesses in the Consumables

  1. Items that will come in the Consumables Package:
  • A micropipettor
  • Primer
  • Buffer Solution
  • A container with plastic grips on the tubes
  • PCR mix
  • SYBR Green
  • dNTP
  • glass slides

The consumables consisted of tubes, the tips, a micropipettor, PCR mix, and primers, but how they were packaged could create a messy workplace. The micropipettor tips were nicely placed in an organized tray, but the tray itself had some design issues. Due to an open bottom and loose enclosure of the tips, they were easily knocked out of the holding tray. And a certain type of micropipettor was needed so that it could fit in the tubes. The package would sometimes include the wrong type of micropipettor that was needed, in order to, pump in the PCR mix and primers. In a real life experiment, this would be an easy cause for contamination. The tubes can easily be knocked over spilling the contents of the samples, which could mix up the data. Not having the right type of micropipettor that fits in the tube could also create a mess at the workbench in lab, if the tip of it gets stuck in a tube and force is used to try and get it out. The consumables will be packaged more closely together to ensure that there are no gaps when delivering the product. And the micropipettor will be inspected to make sure that it is in the correct size that will fit in the tube, in order to suck up the primer and PCR mix to combine them. The tubes will also be reinforced into the plastic tray with a plastic grip that makes sure that the samples made do no come out easily, when making the samples, but still come out with relative ease when the samples need to be moved and tested in the PCR machine.

Feature 2: Hardware - PCR Machine & Fluorimeter

Weaknesses of the PCR Machine & Flourimeter

Although both the PCR Machine and Flourimeter worked successfully in their own ways, both systems had their own different weaknesses. The PCR Machine was very simple and easy to use and resulted in successful end results. However, the machine could only hold a certain set number of samples because of this the samples took a long time to process in the PCR machine. The Flourimeter itself was very portable and easy to set up anywhere, but because of this, the entire system was very unstable. The camera phone that was placed in the cradle moved at the slightest touch and resulted in inconsistent results and pictures. The pictures sometimes came out blurry and slightly unfocused if the cradle was moved out of alignment with the blue LED light on the fluorimeter. Even though the distance of the phone from the fluorimeter was recorded with a ruler, the results are most likely slightly inaccurate because the exact distance between the two was hard to keep track of sometimes and could have gotten slightly moved out of position every time a drop of SYBR Green was mixed with the sample as well as when the light box was moved in order to shut out the outside light.