BME100 s2018:Group6 W0800 L6

BME 100 Spring 2018	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: Dante DeSimone Name: Connor Phillips Name: Jesus Pena Name: Markaya Hunter Name: Aarya McEwan LAMARtech LAB 6 WRITE-UP Bayesian Statistics Overview of the Original Diagnosis System In this lab, the DNA of 22 patients were tested by 11 groups, each group acting as the clinician for 2 patients. The DNA of each of the patients was tested using conventional PCR using Open PCR technology for a certain SNP known to be linked with diabetes. The diagnosis obtained by each group through Open PCR was checked against the “true” diagnosis as given by a doctor examining symptoms of the SNP. A number of steps were taken by each group to ensure that as few errors were made as possible. First of all, each group tested three samples for each patient to ensure that the result was not a fluke. Second of all, each of the three samples for each patient was checked against a positive and negative control that were designed to ensure that the test was working and to ensure that there was some kind of threshold for whether or not a patient should be considered positive or negative. A third way that errors were reduced is that 5 samples of calf thymus DNA were checked for fluorescence to create a calibration curve that, when run through ImageJ, allowed us to calculate an exact concentration of DNA in each of the samples for the patients. Finally, for each sample, three images were taken, meaning a total of 9 images per patient over 3 unique drops were used to determine the final diagnosis. All of these layers of support make the DNA more reliable and trustworthy. The results are as follows, and these results were then analyzed using Bayesian statistics to determine their reliability and validity: Analysis of PCR strengths and weaknesses <tbody> </tbody> Consumables Open PCR Fluorimeter Strengths: Pipette tips are extremely cheap Primer is easily designed/checked online Primer, polymerase, and other chemicals can be mixed/stored in one container Strengths: Accurately facilitated DNA replication Fairly simple to use Lower cost design Connects to a computer easily/effectively Strengths: Beam of light worked well to excite molecules in water droplet Did not light up negative controls, lit up positive controls Very simple/cheap Weaknesses: Each consumable can only be used once PCR mix must be sensitively handled (i.e. temperature controlled, etc.) If you make a mistake (i.e. put something into the wrong tube) you can’t undo it Weaknesses: Requires specialized knowledge to program Takes a long time Heating is inefficient Can only hold a couple PCR reactions at a time Is very expensive Weaknesses: Impossible to both hold phone and close lid of fluorimeter at the same time Image processing is tedious and separate from fluorimeter Does not connect to computer Very difficult to place water droplets on the slide accurately so they stayed in place Patient ID Clinician (Group) CONCLUSION (PCR) Doctor's Diagnosis (no = no disease, yes = disease) 25053 8:00 am Group 1 NEG no 57814 8:00 am Group 1 NEG no 37987 8:00 am Group 2 NEG no 29674 8:00 am Group 2 NEG no 69628 8:00 am Group 3 NEG no 57443 8:00 am Group 3 NEG yes 10452 8:00 am Group 4 NEG no 45121 8:00 am Group 4 NEG no 70416 8:00 am Group 5 POS no 90490 8:00 am Group 5 NEG no 94130 8:00 am Group 6 NEG no 76365 8:00 am Group 6 POS yes 35518 8:00 am Group 7 INCONCLUSIVE yes 69697 8:00 am Group 7 NEG yes 86632 8:00 am Group 8 NEG no 58297 8:00 am Group 8 POS no 55307 8:00 am Group 9 NO TEST yes 75503 8:00 am Group 9 NO TEST no 72874 8:00 am Group 10 NEG no 53428 8:00 am Group 10 POS yes 69222 8:00 am Group 11 NEG no 45886 8:00 am Group 11 POS yes What Bayes Statistics Imply about This Diagnostic Approach Calculations one and two were to test the PCR system's ability to measure the presence of a disease SNP. The results show that the PCR test was a reliable experiment to determine whether a person has the disease SNP or not, both the calculations were close to 1.00 (100%). Calculations three and four were meant to test the reliability of the PCR system to predict the development of the disease compared to that of a doctor's diagnosis. The results for calculation three show that the PCR system was not reliable in determining whether or not that person has the disease SNP. However, calculation four proved that the PCR system was reliable at determining whether or not the subject does not have a disease SNP. There were many places were error could have accumulated in this experiment. The micro pipetting process was unable to transfer all of the necessary liquid. This could skew our results based on the amount of replicate PCR in the sample being tested. Given the elementary nature of our picture taking set-up, there is a high probability that the image taken was not a 100% accurate representation of the result. This would change the results obtained from processing the image and ultimately our results. This of course is under the assumption that the images were processed flawlessly. Given that the picture processing requires a lot of human estimation, it cannot be said they were processed entirely without a margin of error. Intro to Computer-Aided Design 3D Modeling The software that the team used was SolidWorks. The entire team has used SolidWorks prior, in the class BME182; so we all had some knowledge of the software. However, this exercise was not as intensive as compared to the work done in BME182. Most of the knowledge that we learned in BME182 helped us out immensely , most notably the knowledge of "mating" parts. While we did not go in detail what each mate does, we knew the general idea of it. With the general knowledge of mating and our intuition, we found how to mate all the parts perfectly, and it did not take a long time. That being said, we most likely did not even use 1% of the capabilities of SolidWorks. So even though we figured out the exercise, SolidWorks is a very tricky software to use because of it's capability to do so many different tasks. Our Design We chose this design because it solved the 2 major problems the original fluorimeter has. Our different from the original design because it is more reliant on a software than humans. This helps it minimize the problems we had, which were the amount of light entering, and the consistency of the photos. Rather than having a human run the experiment, it will be more computerized and automatic. A human will almost never take the same photo every time, same height, same distance, etc. Since our camera is in a fixed spot, this helps take more consistent photos. Feature 1: Consumables Consumable Kit:     Since the consumable kit is just a kit, there cannot be a lot of consumables inside the box or else it would be inconvenient and ironic. So rather than having a computerized and automated assembly, it will have to include the necessary consumables and a few other things to minimize the size of the kit. So other than having the necessary tools, like primer, pipette and polymerase, we will include containers with specific requirements labels. The labels will have requirements like temperature, the amount of primer, the amount of polymerase, etcetera. Not only will they have requirements, but they will also have step by step  This will help the people doing this test to be as accurate as possible. Also, to help out the people, we will package the tools in such a way that you do not have to look around for the tools you need. Rather, they will be already placed next to each other for less confusion and convenience. Feature 2: Hardware - PCR Machine & Fluorimeter PCR Machine: The new machine will be more computerized to minimize the amount of human interaction. In the machine there will be a port where the phone will be placed. It will be leveled correctly so that the image can be seen. Since the machine will be more computerized, instructions will pop up on the screen to tell the user what to do step by step. The system will also be able to hold more data so that the user can store more PCR reactions. Fluorimeter redesign: The new fluorimeter will be much more integrated into one design than our original design. While the box and phone design is very simple and cheap and thus accessible, it has a number of flaws that make it less effective. First of all, the phone is not able to properly stand and take a picture of the water droplet while the door is closed. This means that the user is forced to choose between the ability to take a picture accurately and the ability to darken the area around the droplet to provide better contrast in the picture. Secondly, because the fluorimeter setup is not connected to a computer, the process of file management and analysis with the images becomes difficult and tedious. The user must transmit images from the phone to computer in such a way that each image is meticulously labeled-- it is far too easy for the user to make an error in this step alone. The image analysis that follows from this is very tedious as well, as the user must open each file in Imagej and then manually edit and analyze the image. With these two weaknesses as our guide, we have decided to make a few modifications to modernize and improve the fluorimeter setup. The first and most important change is to add a camera directly to the wall of the fluorimeter. This camera would be embedded in the door of the fluorimeter such that when the door is closed, the camera will be fixed at a set height above the ground. This addresses the weakness of the phone in two major ways: first, it ensures that the fluorimeter setup must be darkened for a picture to be taken. In addition, it ensures that the camera is stable and at a set predictable height so it is very easy to position the droplet accordingly. This camera will take a picture on the command of either a physical button or the control of the software described below. This software is the second major update to the fluorimeter. This software that will interface with the fluorometer and a computer. This software will be modeled in some ways off of the Real Time PCR software in that it will work on any computer if you type in the IP address. This allows it to be flexible and effective. Using this software the user can pictures and group images into folders in real time. This eliminates the danger of mislabeling or forgetting certain files. The software will also be preprogrammed to ready the image for imageJ analysis so all the user has to do is draw the circle and collect the data. These two modifications will allow the design to be as user friendly as possible and thus effective, while not adding an extreme amount of cost.