BME 100 tested 34 patients for the disease-associated SNP. There were 17 groups of up to 6 students each and every group was assigned two patients. The class performed a PCR reaction on positive/negative controls for each group, and three replicas of each patient. To perform the reactions, 50 μL of PCR reaction mix was combined with 50 μL of either control or patient DNA. Then, this mixture was placed in a PCR machine. To avoid errors in differentiating the samples, each tube was carefully labeled and recorded in openwetware(lab4). To avoid errors in contamination, fresh pipette tips were used with every separate use of the pipettes. After the PCR machine was run we analyzed the presence of the DNA SNP with a fluorimeter system and the imajes we took on smart phones. To avoid major differences in camera angles and positions we measured out 4cm between the phone cradle and the camera before taking every picture.To make sure our ImajeJ analysis of our pictures was accurate, we used calibration samples to determine a line of best fit. To make the picture data more reliable, we took three pictures of each dot containing the SYBR Green and the sample. The averages of the three photos were used. Our final data from all of the groups was compiled into one spreadsheet for final analysis of our test. This spreadsheet contained mostly successful conclusions, but did have a few wrong, and a few inconclusive.
What Bayes Statistics Imply about This Diagnostic Approach
The probability of a positive result in our BME 100 test actually corresponding to a sample with positive DNA Sequence for SNP, was close to 100%. The probability of a negative outcome in our BME 100 test actually coming from a negative sample, was slightly less, but also close to 100%. This means that our test was pretty good in terms of reliability.
Given a positive result in our BME 100 diagnostic test, the patient has about a 50% chance of developing the disease SNP. If our BME 100 test signals that the test is negative, the patient has almost 100% chance of not developing the disease SNP. This means that the test is better at assuring that someone will not get the disease, than it is at predicting that they will develop it.
One of the major possible errors is the inconsistent placing of the phone distance from the drop. The distance needs to be kept the same throughout the entire process otherwise the measurements will be different estimates for each drop. It was a difficult task to ensure the phone was placed exactly in the same position every time a new drop was analyzed. Another error that could occur has to do with the lighting. If the box is not utilized correctly, the surrounding room light will interfere with the light wavelength while the molecules on the droplet could be flourescing. Although usually not as common, another issue is making sure that the phone camera focuses on the actual droplet to make sure blurry images are not used. The oval shape is hard to draw around a droplet if the photo uploaded on ImageJ shows blurry spaces around the drop.
Intro to Computer-Aided Design
3D Modeling
Our group decided to work with the software Autodesk Inventor and Assembly. One of our team members has past experience with this program, so we decided on using it to make the design process faster. Our team was able to clearly put our ideas into the program, without spending large lengths of time struggling with a not so familiar software. Autodesk Inventor is much like SolidWorks, but our group decided on using Autodesk simply because the person designing felt more confident with this program.
Our Design
Design using Autodesk Inventor of new box with hinged side.Design using Autodesk Inventor of new pipette box that on top will contain pipette tips to the right side, and the PCR solutions labeled and ready to be used. Below the pipette tips, there will be a slot for the glass slides used for the PCR drop analysis.
Our design now consists of a fully closed, dark box. Only one of the sides will be able to be opened, which will use a hinge to left the side up, in order to insert the phone and the glass slides. This will provide total darkness inside the box, besides the blue drop going through the drop. This will provide for better pictures, therefore better pictures to analyze on ImageJ. Our phone cradle was also improved. This new phone cradle provides for better angles, phone stability, and a quicker process for taking the pictures. The height of the cradle can be adjusted using a lever motion in order to level the phone camera with the droplet, so that they are almost directly parallel to each other.
Feature 1: Consumables
The current consumables that were used in the lab had weaknesses that included how the materials were packaged and distributed, and the wastefulness in the act of micropipetting. To save a lot of space and make the process much neater, we propose that all of the materials (the micropipetter, the micropipette tips, the glass slides, the PCR mix, the primer solution, and the SYBER green) would all be included in the same package. Modifications to the micropipetter and the micropipette tips also needed to be made to reduce the wastefulness of this specific lab. Instead of using a 100/1000 μL micropipetter, a 20/200 μL micropipetter should be used. Instead of using 1000 μL micropipette tips, 200 μL tips should be used.
Feature 2: Hardware - PCR Machine & Fluorimeter
Some of the major weaknesses that we found with the process were based around consistency and clarity with the Fluorimeter. Every time that our group was done with one slide and had to move to the next, the stand that held the smart phone would inevitably get moved. This resulted in slides that had pictures taken from different angles, therefore making each set of three pictures per slide look different from the others. Our idea is to make the box that contained the fluorimeter system completely closed by adding another side so that the only time that light enters the system is when hinged side is open. The added side would be the bottom, or floor, of the system. Attached to the added side would be a cradle, "phone stand", that firmly holds the phone so that there is no way the pictures could be inconsistent. Additionally, the cradle will be able to adjust the height that the phone takes pictures out, but still holds the phone firmly. We also found that the problem of picture clarity could be resolved by taking design elements from the infamous "selfie-sticks". A feature of some of the newer "selfie-sticks" is a device with one button that connects to the phone using Bluetooth. Incorporating a device like this into the Fluorimeter would greatly improve the clarity of the pictures because the box could be completely closed while taking the pictures, removing any outside light that might interfere with picture clarity.
BME 100 Fall 2017
