# BME100 f2015:Group13 1030amL6

BME 100 Fall 2015 Home
People
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
Photos
Wiki Editing Help

# OUR COMPANY

 Name: Drew Worman Name: Tanner Ivey Name: Bramuel Simiyu Name: Christopher Chen Name: Jonah Brosemann Name: Kaylee Antill

# LAB 6 WRITE-UP

## Bayesian Statistics

Overview of the Original Diagnosis System

The BME 100 classes tested the DNA of patients for the disease associated with SNP. There were 17 groups of six students each. Collectively these groups diagnosed a total of 34 patients using am OpenPCR system including an Open PCR machine and a fluorimeter. There was a total of eight different experimental DNA solutions that were tested for the disease per patient. This meant that in order to prevent error, there were three photos taken per solution totaling in 24 total images per patient that were analyzed using Image J. These PCR machine and the Image J software also had controls of their own. The PCR machine kept the DNA at a consistent temperature cycle for all the samples at once to prevent any variation in that aspect. The controls in the Image J calibration were that when we analyzed the three images for each PCR sample, we used the black background of each image to act as the base for the analysis for that particular image. Out of all the class data, there were sixteen successful conclusions that correctly diagnosed the patient. This left fourteen incorrectly diagnosed patients, two inconclusive tests and two untested patients. Despite all the controls set in place to avoid any potential error, there is still room for it. One major challenge that affected the outcome of our data was that the camera on the iPhone were were using was not focusing easily. So many of our photos came out slightly blurry. This would affect the outcome of the picture analysis and thus the outcome of our diagnosis.

Calculation 1 describes the probability of a positive test conclusion, given a positive PCR reaction. A high probability close to 100% means that a positive PCR reaction test result is an accurate measure for a positive final test conclusion. Oppositely, a low probability means that a positive PCR reaction test does not reflect a positive final test conclusion. Calculation 2 describes the probability of a negative final test result, given a negative diagnostic signal. A high probability of close to 100% means that a negative PCR reaction test will likely coincide with a negative final test conclusion, while a low probability shows that a negative PCR reaction test is not an accurate measure for a negative final test conclusion.

Calculation 3 describes the probability of a patient developing the disease given a positive final test conclusion. A high probability close to 100% means that the test conclusion is an accurate measure of diagnosis for someone who has the disease, while a low probability means that a positive final test conclusion does not mean that the patient will develop the disease. Calculation 4 describes the probability that a patient will not develop the disease given a negative test result. A high probability close to 100% shows that the test is an accurate measure of diagnosis for those who do not have the disease, while a low probability shows that a negative final test result does not coincide with a patient not developing the disease.

Possible sources of error can include improper operation of the micropipette. This can result in not all of the PCR reaction sample or not all of the SYBR Green being pipetted onto the slide. This can result in a misdiagnosis when using fluorescence detection. Another source of error can come from not disposing micropipette tips after a single use. This will result in the mixing of multiple DNA samples when using the micropipette, which in turn can lead to a patient being diagnosed for a disease they do not have or not being diagnosed for a disease they do have. Letting light into the fluorimeter when capturing images of the sample can cause improper fluorescence analysis of the images. All of these errors can adversely affect Bayes values.

Calculations

Calculation 1

 Variable Description Numerical Value A Positive final test conclusion 0.41 B Positive PCR reaction 0.45 P(B|A) Probability of positive PCR reaction given positive final test conclusion 0.85 P(A|B) Probability of positive final test conclusion given positive PCR reaction 0.77

Calculation 2

 Variable Description Numerical Value A Negative final test conclusion 0.53 B Negative diagnostic signal 0.49 P(B|A) Probability of negative diagnostic signal given negative final test conclusion 0.82 P(A|B) Probability of negative final test conclusion given negative diagnostic signal 0.89

Calculation 3

 Variable Description Numerical Value A Patient will develop disease 0.31 B Positive final test conclusion 0.41 P(B|A) Probability of positive final test conclusion given the patient will develop disease 0.31 P(A|B) Probability that patient will develop disease given postive final test conclusion 0.24

Calculation 4

 Variable Description Numerical Value A Patient will not develop the disease 0.69 B Negative final test conclusion 0.53 P(B|A) Probabiity of a negative final test conclusion given the patient will not develop disease 0.71 P(A|B) Probability that patient wil not develop disease given negative final test conclusion 0.91

## Intro to Computer-Aided Design

TinkerCAD is an easy to use 3D modeling tool which can be accessed through web browsers. We used it to model parts of a PCR machine and fluorimeter as well as explore the modifications that we will be making to the machine and/or fluorimeter. In particular, we were interesting in modifying part of the PCR machine to create more slots to hold samples. Overall, learning to use TinkerCAD was a much quicker and more intuitive process than using Solid Works.

Our Design




First we had to decide what exactly we were going to model. We decided to improve the subject tray so that it would be capable of testing more subjects and getting more results. First we made a base of similar dimensions and then enlarged it about 150%. We then added an additional 20 slots to more than double the output of the experiment. This could slightly affect the time needed to process the samples. We saw this difference in time as being negligible considering the far greater amount of samples that could be tested with the machine. In situations in which time is not a major constraint, our improved machine could actually save the user a large amount of time. For example, with the original PCR machine if a user wanted to test 30 samples overnight, they would either have to test the samples over a span of two nights or use two PCR machines. Our device would allow them to test all the samples in only one night using a single PCR machine.

## Feature 1: Consumables

When it comes to consumables, some are considered more important than others. "Very important" is a term used to describe the consumables that can not be replaced by any other devices and are essential to the product. Our kit will not contain any new materials because our design only addresses the amount tubes that are able to be placed in the machine and the placement of the user screen. Based on this, in our kit, we will have:

- Plastic Tubes
- PCR mix
- Primer Solution
- SYBR Green Solution
- Buffer
- Micro-Pipette Tips
- Micro-Pipettor

These items will enable the user to perform several tasks with one kit.

## Feature 2: Hardware - PCR Machine & Fluorimeter

In our overall system ,consisting of a PCR machine and fluorimeter,the fluorimeter will be utilized in the same manner for which it is currently used (allowing for the photographing and reading of samples). It will not be modified in any way. The PCR machine will be slightly altered to make it more useful to users. The Open PRC machine is a fairly cheap, easy to put together device that can be used in a wide variety of settings. In some of these settings, PCR might need to be carried out on a great number of samples. The current PCR machine only has 16 slots for inserting samples, which could greatly limit the rate at which samples can be processed or necessitate the purchasing of another PCR machine. This would negate some of the current benefits of the PCR machine, however (namely the low cost). We would redesign the PCR machine to have 36 slots for samples, allowing more samples to be tested at a reasonable rate. This would slightly increase the size of the machine, limiting portability. We believe that this compromise is still a much better result than having to purchase more machines or having the rate of sample processing greatly reduced.