BME100 f2016:Group6 W1030AM L6

BME 100 Fall 2016	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: Matt Edick Name:  Edgar Manriquez Name:  D. Brennen Martin Name: Megan O'Reilly Name: Blake Stephens LAB 6 WRITE-UP Bayesian Statistics Overview of the Original Diagnosis System Patients were tested for disease associated SNP using a PCR system and fluorimeter. 15 teams within the class measured 2 patients each, and used positive and negative samples as controls. In order to ensure that the results were accurate, each patient's sample in the fluorimeter had 3 pictures taken. By doing this, when processed in ImageJ, the information would have several replications therefore allowing for the mean to be taken. This ensures a higher level of accuracy within each group's results. In ImageJ, it is simple to ensure that the space measured per drop is the same using the measurements previously recorded, a feature available within the program. The statistics were then measured on an excel sheet in which all data was recorded and turned into dot plots and was easily manipulated for means and any other desired statistics. When the class's results were compiled, two groups did not make their data available, resulting in only 26 samples to be used for conclusions and statistics. It was concluded that 87% of positive results as concluded by students were actually positive, 94% of negative conclusions were actually negative. There were no inconclusive results as found by the groups which was correct as known from the samples. This shows that overall, the data collected and the conclusions made from those samples proved to be accurate when compared to the known conclusions for the samples. Within our own data collection, there was a slight difficulty in the labelling of the samples which then resulted in us needing to recognize the trend in our data from ImageJ and reorganize our data. However, once the data the recorded and organized, our conclusions matched up to the known conclusion of the sample. What Bayes Statistics Imply about This Diagnostic Approach Calculation 1 and 2 refer to the detection of the disease by the system which has a high reliability since both calculations are above 80%. Calculation 1 refers to the sensitivity of the system to detect the disease SNP which has a moderately high reliability, about 80% accuracy for the 26 samples in our class. Calculation 2 describes the specificity of the system to detect the disease SNP which has a very high reliability, nearly 100% accuracy. Calculation 3 and 4 refer to the prediction of the development of the disease. Calculation 3 refers to the sensitivity of the system to predict the development of the disease and has a moderate level of reliability, nearly 50% accuracy. Calculation 4, which refers to the specificity of the system to predict the development of the disease, is more reliable, nearly 100% accurate. There are several possible sources of error that could have occurred which could have affected the Bayes values in a negative way. Specifically, human error could have led to incorrect measurements throughout the process including during the usage of the micropipettor to the calculations of the sample statistics. Another possible source of error is with the labelling of the tubes which then would affect the organization of the data throughout the entire process of recording and manipulation the sample data. During the lab, there was a miscommunication between the labelling of the tubes and pictures during fluorimetry which caused us to double check the data that we had and adjust accordingly. A third source of error is within the ImageJ system which seemed to have some fluctuations within the data, showing that it might not be completely accurate. It is important to note that within the system, making sure that the circles around the drops had to align for each measurement in order to ensure a high level of accuracy. However, even with this accounted for, there were some instances in which we measured the same image twice but got different results. This could have affected the results that we had and therefore had a trickle-down effect, which would then change the results for the statistics and group as a whole. Intro to Computer-Aided Design 3D Modeling Our Design Feature 1: Consumables There are various consumables that with small adjustments and additions they'll work very well with our improved machines. To come with the PCR machine will be trays that can hold the same amount of plastic tubes the PCR machine can heat at a time. This tray allows the end user to load all their tubes in a tray to be heated outside the PCR machine and be able to label directly on the tray and organize their samples better all without effecting the function of the PCR machine. All the fittings are standard with universal fit eppendorf tubes. The great feature about this tray is that it isn't required to load the PCR machine, but is included more for convince sake and better organization practices. There are no other consumables we can provide with the PCR machine that the user wouldn't get themselves like primers and SYBR green. The same goes for our improved fluorimeter, the consumables don't need a make over but added features to work better with our machine for convince. Since our fluorimeter is a completely closed system it makes adjusting the slides by hand tricky, therefore we have included a nob which moves the slides from the inside for changing position and making adjustments that are fine tuned for use. Our device will ship with slides that have grooves on the bottom to grip the fine adjustment better than smooth bottom slides and allow for confident adjustments. In addition to these unique slides, our internal camera is just that internal. So there needs to be an interface between the machine and a computer to export hte photos. Which is why we will include a usb cable to go along with the machine. Feature 2: Hardware - PCR Machine & Fluorimeter The PCR machine currently has several weaknesses that can be improved upon to make the machine more efficient and easy to use. Our design will have a larger screen and user interface that allows for the set-up to be done on the PCR machine's screen rather than requiring a separate monitor to program the cycle information. The wooden siding on the bottom of the machine will also be replaced with plexiglass which will make it easier for users to see in the machine. This will make the machine more ideal for educational scenarios as students will be able to see PCR in real time. Aesthetically, the addition of our logo laser engraved into the plexiglass surface could make the appearance of the product more pleasing for users. The fluorimeter also has a few issues that our design improves upon. The machine will be enclosed with a small sliding door accessible on the side which will allow for a more efficient machine. By not requiring the user to lift up and reorganize the slide each time, the machine will be easier to load and use. Another problem is the need to set up a camera externally and set up a timer, as well as settings. Our design has a fixed internal camera which will eliminate a lot of hassle for the user. Additionally, there will be a USB port which allows for the pictures taken, by the internal camera, to be exported to a laptop, phone, or external monitor for analyzation.