BME100 s2015:Group2 9amL6

From OpenWetWare
Jump to navigationJump to search
BME 100 Spring 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


Ryan Bridges
Name: Michael Bejarano
Name: Virginia Fernandez
Morgan Johnson


Bayesian Statistics

Overview of the Original Diagnosis System

BME100 tested patients for the disease-associated SNP by using PCR. There were 8 lab groups in the morning section. Each lab group had to diagnose two patients. Each patient had three samples that were individually tested for the disease-associated SNP. Each sample was put in a small test tube and PCR reaction mix along with DNA/primer mix were added in order to prevent error, a new micro pipette tip was used for each addition. In addition, there were three different samples for each patient in order to minimize error. Positive and negative controls were used as a reference to confirm the result. Three separate images were analyzed for each sample to insure that the results were accurate and to minimize error. The BME100_fa2014_PCRResults spreadsheet showed that there were 14 successful conclusions and 2 inconclusive results. Group 5's data was eliminated because it was incorrect. Our group changed one of the inconclusive conclusions to negative because two of the samples tested negative and only one tested positive.

What Bayes Statistics Imply about This Diagnostic Approach

The results of calculation 1 implied that the ability of the PCR system to correctly detect the presence of the disease SNP was remedial. The system only had somewhat hogh chance which is fairly close to 1.00 but not close enough for the system to produce extremely reliable test results. The results of calculation 2 implied that the PCR system was very accurate at producing negative final test conclusions. The probability was slightly larger than 1.00 which means the the results are not completely reliable, but the probability is close enough to be considered mostly reliable and accurate. The Bayes values could have been affected by error such as inaccurate data due the continuous distance change of the smart phone (or camera) after each new droplet smaples was placed; overall errors with ImageJ software is believed to be minimal. Errors were prevented with ImageJ calibration software by setting the parameters correctly before proceeding. The steps involved the selection setting measurements, and choosing “area integrated density” & “mean grey value.” Once the image was loaded to ImageJ the options selected follows as:1) image 2)color 3) split channels and 4) keep the green window. Each drop had a total of three pictures taken to be analyzed, and each picture’s “green” split channel was analyzed. The true error reduction occurs with the next procedures in the analysis of the green image. Select manger window and then continue with selecting the task bar’s oval shape tool. Using the oval tool overlay it the on the darkest 'center' spot on the droplet and then select analyze. The analysis will bring up a value of the area in the manger window, and this area value can be used again to the following two duplicate pictures. Again repeat manger procedure when analyzing the background values. These procedures greatly reduce ImageJ errors within our analysis. In addition, the samples that were being tested could have been contaminated if a new micro pipette top was not used each time the solution was transferred from one container to another, or if the containers were left open and a contaminant entered the solution. Furthermore, if the SYBR green was left in the light too long, it could have possibly gotten bleached and not turned green in a positive test solution.

The result of calculation 3 implied that the reliability of PCR for predicting that the disease would develop was somewhat reliable. The value was closer to 1.00 than 0.00 so it was considered to be reliable at diagnosing the patient most of the time. The results for calculation 4 implied that PCR was also somewhat reliable at predicting that the disease would not develop. The value was close to the value of calculation 3 so PCR was considered to be reliable at predicting that the disease would not develop the majority of the time.

Computer-Aided Design


      TinkerCAD or Tinker Computer Aid Design is used to aid the user in designing products, and provide animation to give the user an idea in
how it appears. This Computer Aid Design program was used to construct a polymerase chain reaction machine for are group. We the provided
templates provided by Dr. Haynes we were able to construct a modified polymerase chain reaction machine. In addition the computer
aid design program gave us an idea on how the light will travel through the outlining walls. TinkerCAD is a very useful tool
for the user to visualize how a product will appear.

Our Design

The original design of the polymerase chain reaction machine was rather effective, expect for one frustration which was having to prop the smart phone consistently. To resolve that issue a place holder for the smart phone was planted inside the polymerase chain reaction machine. This feature will benefit the user by no longer having to struggle with propping up their smart phone continuously. The user can place the drop the DNA simple inside the polymerase chain reaction machine via top primary open slot. The position of smart phone holder was concluded based on how light will travel through the polymerase chain reaction machine, and the believed optimal location to reduce glare.

Feature 1: Consumables Kit

The elements of the consumables kit must be packaged and handled in a careful manner to eliminate any possible weaknesses or errors that may result from the mishandling of the kit. The packaging will ensure that proper cushioning material, preferably within a strong outer container, is being utilized to protect the components of the kit from damage. Several factors, especially temperature and handling, need be considered when packaging the kit as these factors can directly affect the effectiveness of the kit. The liquid reagents will be properly insulated to avoid exposure to extreme temperatures, using fiberglass insulation material or styrofoam to maintain a steady temperature. All materials will be packaged and cushioned individually to best prevent damage within the kit, including the micropipettor, tubes, and tips, while the pcr mix and primers will be well-sealed to prevent any external contamination. Taking these measures will reduce the effects of external factors damaging the consumables kit.

Feature 2: Hardware - PCR Machine & Fluorimeter

The fluorimeter is attached to the PCR machine in a position that would be most effective for illuminating the sample where the least amount of light is present inside the box. Of course this may vary, so the fluorimeter would be readjustable within the PCR machine.

Although the PCR machine and fluorimeter used in the lab was effective in blockading light and placing the LED light in such a way to highlight the samples in the drop, it was an issue to have to constantly focus the camera and readjust the phone. In order to take the clearest picture of the DNA in the drop, our new fluorimeter design would have a stand for the phone on the inside that would be easily adjustable, yet stable so it will not move and lose focus when the box is closed. Not having to constantly readjust the phone position according to the system would save time, due to the design of the new box compared to the old one as well as the stable phone stand. This would enable in clearer pictures and better analysis using the ImageJ program.