BME100 f2015:Group10 1030amL6

From OpenWetWare
Jump to navigationJump to search
BME 100 Fall 2015 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


Gerrit Orthlieb
Sergio Mejia
Name: Brandon Burgers
Sai Adoni


Bayesian Statistics

Overview of the Original Diagnosis System

17 groups of students were separated in order to test 34 patients for a disease caused by a single nucleotide polymorphism (SNP). Meaning each group tested 2 patients each. 3 samples were taken from each individual in order to minimize error in the OpenPCR procedure. However, prior to the testing of the PCR, a calibration curve was made by using a program known as ImageJ to get samples that had forsake DNA present. A positive and a negative were used in tandem with the patient samples during the PCR in order to ensure that the methods that were used were accurate and done properly. Three images of each sample were taken in order to make sure that there was very little room for error.

The final results showed that 32 of the patients had conclusions, and the other two patients were blank. From the conclusion results, 17 patients were positive, 13 patients were negative, and two patients had inconclusive results.

From frequencies, our group specifically showed that the data was relatively consistent. One problem our group seemed to have was not being able to leave the phone in the phone stand without the stand falling over. It seemed to be too top-heavy. Also, when writing on the samples, we found that it was quite difficult to write on them without it smearing or even showing up on the label.

What Bayes Statistics Imply about This Diagnostic Approach

Calculation 1 dealt with the probability that a positive final test conclusion was determined given that a positive PCR reaction occurred. This is demonstrative of the reliability of the individual PCR tests based on whether or not they corresponded appropriately to the final conclusion of if the patient had the disease or not. Given that the probability of a positive final test conclusion given a positive PCR reaction was around 77%, it shows that a single PCR test is mostly reliable in determining a final conclusion of the disease and should be used tentatively as a final deciding factor when considering a diagnosis. Calculation 2 agrees with this point, as the probability of a negative final conclusion given a negative PCR test is around 88%. The reliability of a single PCR test being able to determine a final conclusion is fairly large, but when dealing with patient lives, multiple tests should be run to ensure validity of confirming or rejecting the hypothesis that the patient has the disease.

For calculation 3, the probability of the patient developing the disease based on a final PCR test conclusion is around 30%, a particularly low value for a test that is supposed to determine the presence of a disease. There is a fairly low reliability of the PCR final conclusion being able to predict the development of the disease. Also, calculation 4, which dictates the probability that a patient with a negative PCR final conclusion will not develop the disease is 79%, a much higher value, although nowhere near ideal. The final PCR conclusion tests would have to be run seven times with all positive results in order for a positive test result to mean more than 90% chance of developing the disease. The final PCR conclusion test would have to be run only two times with both negative results to have greater than a 95% probability of the disease not developing.

There are, however, possible sources of error that may account for this unreliability. Due to the small volumes being dispensed, it is possible that not all of the dye or sample was conveyed to the PCR tube for reaction and may have adhered to the sides of the PCR tube instead of reacting with the primers and dye. Another possible source of error includes the shielding of the sample when taking pictures for use in ImageJ. It is possible that when repeating the picture-taking process, the black box may have been sealed differently as to let more or less light in and skew the results of the color analysis. Finally, it is possible that the camera may have been positioned at different distances from the sample when the fluorescence images were being taken and could have altered the ImageJ results in that manner as well.

Intro to Computer-Aided Design

Maneuvering through the TinkerCad design process required close attention to detail and close reading of instructions provided in the lab workbook. When designing the new PCR machine it was helpful to upload two designs that were provided by the instructor on another site. However, the lab group decided to make changes to other devices that would improve reading for PCR machine. The first object that was designed was a phone stand that could hold any phone size without tipping over and be able to capture a more precise picture. The next object designed were small sticker strips that could be placed on test tubes to accurately identify tubes and reduce chances of marker smudging on tube. TinkerCad made it possible by being able to select different shapes and orientations to design any desired object for the lab.

Our Design

The design for the phone stand is different from the phone stand provided for the lab because there is no restricted compartment. A phone can be placed of any size to capture a clear picture once placed in the box. The phone stand is also supported at two ends to prevent the phone stand from falling over and a wheel is placed at one end to adjust length of area where phone is placed. In this visual, there is also small sticker strips that were designed to be used for labeling test tubes.

Feature 1: Consumables

The products used for consumables will be micro-pipet tips, reagents, and PCR solution mixes. The PCR consumables kit could be improved by including special labels on the PCR solution mixes. This allows for a more careful experimental procedure. The labels and pipet tips could be heat resistant so in the event that the pipets come into contact with a high temperature, there is little to no risk of any harm and the sterilization process will be a lot easier. The pipets will be manufactured with state-of-the-art durable material which makes it very useful for multiple experimental runs. The pipet tips used can be safely and easily disposed. The overall consumables kit is fairly cheap compared to other consumable kit brands while being efficient at ensuring that the lab's experiments run smoothly and hassle free.

Feature 2: Hardware - PCR Machine & Fluorimeter

Both the Fluorimeter and the PCR machine are included in our design. In the original experiment, we had difficulty photographing the sample because of the angle of the camera and because the camera kept falling down. In order to solve this, we did not make changes to the PCR machine or the fluorimeter, but to the camera stand itself. In order to solve the problem of the camera falling down, we made the camera stand taller, and we also added a knob on the side in so that we could adjust the size of the camera stand to fit any type of camera. Other than that, we had minimal difficulty obtaining images of our samples.