The BME 100 class has 17 teams of 6 people each diagnosing 32 patients. Each group has two different patients that no other group has. The group members wore gloves and a lab coat to prevent their own DNA to contaminate the patient DNA. The group members made sure the fluorescence was out of the light as much as possible. The group used about 80 μg of the patient DNA and the solutions, leaving about two trials per mixture. Each group had a negative and a positive control without any patient DNA. One thing was done to prevent error in the ImageJ calibration controls was to reread the directions each run to make sure there was no way possible that there was a mistake. There were 3 pictures per PCR sample to help with accuracy. Out of 96 total PCRs, how many tests per person, 36 were positive and 56 were negative. Out of the 32 conclusions, 12 were positive, 18 were negative, and 2 inconclusive during the DNA testing. Out of the 32 total diagnoses, 9 had the disease and 23 did not. Only 6 had positive results and had the disease, which is about half. Only 16 had negative results and did not have the disease, which is pretty accurate. The group might have touched the tips on accident. Taking pictures at the right time and getting the camera to work was difficult. Keeping the fluorescence out of the light was very difficult.
What Bayes Statistics Imply about This Diagnostic Approach
Calculation one and two were testing how accurate the positive and negative tests were. The group used the data to figure out the percent of final positive or negative (depending on calculation one or two) tests, positive or negative PCR reaction, and the probability of a positive or negative PCR reaction giving a positive or negative conclusion. Using the formula given, the group calculated the probability of a positive or negative final test with a positive or negative PCR reaction. For calculation one, the positive test was close to 1.00 or 100%, which means it is very accurate. For calculation two, the negative test was close to 1.0 or 100%, making it very accurate.
Calculation three and four were testing the accuracy of a positive PCR reaction with the disease diagnosis. The group used the data to figure out the percent of a yes or no (depending on calculation three or four) final conclusion, a positive or negative PCR reaction, and the probability of a positive or negative PCR reaction given a yes or no test conclusion. The group used the formula to calculate the probability of the patient developing the disease or not given a positive or negative final test conclusion. For calculation three, the probability was around 50%, which is not very accurate but still can find the disease. For calculation four, the probability was about 114%, which means the device is very accurate in knowing the patient does not have the disease.
Since the lab is dealing with DNA, the experiment is easily contaminated and is microscopic. One of the possible sources of error was the group members mixed their own DNA with the patient's, on accident, by touching the tip of the pipette or touching their glove with a bare hand and then touching the patient DNA. Another possible error was leaving the fluorescence out in the light, making it react with the light and skewing the data. The last possible error was the inaccuracy of cameras or using screenshots from a video and then calculating the data on ImageJ.
Intro to Computer-Aided Design
TinkerCAD
A majority of us had used Solidworks to make projects in the past. Therefore, the initial thought was to use Solidworks but as the group started to make the design, it looked like it will be too hard to do. TinkerCAD was suggested in the manual, therefore, some of us decided to see what they can build from that. TinkerCAD was easily to use because it had an easy layout and was generally user-friendly. While Solidworks was not user-friendly and was very slow, while being annoying to navigate. Even thou TinkerCAD does not have the same detail, it was better for our project.
Our Design
The design of the PCR machine does have some differences than the one we used in the Lab. One of the big problems that we wanted to challenge was to make the device smaller than it was. Consequently, the device will save more room and even make the process go faster because of the size. There was a added power button, which has the big start letters in front of it, because it would be easier to use than before. We choose this design because we wanted a machine that will work easier and have flexibility moving the equipment around.
Feature 1: Consumables
As a team, we all thought that the capsules were hard to work with. Some of them members thought that the capsules were too small to place the liquid in. This caused some stress with putting the liquid in, sine the person did not want to spill all over. Or members had trouble placing the liquid in because the capsules could easily move around in your hand when they are being transferred in the further experiment. A way to fix both of these would be to have a better container that could hold everything nicely and easily, therefore, people would not be stressed out. Also since the design of the PCR machine will be smaller than before, the capsules would have to be slightly smaller to make sure they can fit into the machine.
Feature 2: Hardware - PCR Machine & Fluorimeter
The PCR machine that we designed was smaller in size due to the lack of portability in the one we used for the lab. Also the size would make the device smaller and easier to use, which might make the process go faster than it did before. There was a big power button added to it so it easier to use.
The fluorimeter that we used had a lot of problems that we faced. One of the main problems was that the picture would not always be in the same place since the camera stood on a platform that could have been moved. Therefore, we decided the stand should be connected to the fluorimeter. This will help keep the image the same size each time. Also the trying to take a picture while the phone was under the box was difficult. Hence, we decided to have a panel that could be taken out easily so the picture can be taken in the dark and not be difficult for lab members to do.
BME 100 Spring 2016
