BME100 f2014:Group26 L6

BME 100 Fall 2014	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 Ambike Bhraguvanshi Timothy Chen Andrew Polson Rachel Ponstein Rebecca Schiavone Jiaqi Wu LAB 6 WRITE-UP Bayesian Statistics Overview of the Original Diagnosis System In this experiment, 68 patients were tested for a disease marker to predict their likelihood of developing the disease. These patient samples were divided among 34 teams in the BME 100 class. Each team, which consists of 6 members, received three replicates of each of the patient's data and two controls. Within the lab groups, the students worked together to perform image analysis. Errors were attempted to be prevented through multiple avenues, including: quantitative results and controls. The replicates of each patient's DNA prevented error, because it allowed for more quantitative data. In addition, there were PCR controls (+/- control samples) that were used in ImageJ to calibrate. The ImageJ calibration was based on the controls. The controls allowed for a baseline to refer to when analyzing patient data. To ensure accuracy, three drop images were analyzed through ImageJ calculations for each patient, and for each control. For the final data, of the 68 tests run, 8 of the results were inconclusive, 6 of the results were blank data and the remaining 54 results led to successful conclusions. What Bayes Statistics Imply About This Diagnostic Approach For Calculation 1, the probability result reflected that the Bayes value is close to 1.00. For the second Calculation, the probability also reflective of the Bayes value being close to 1.00. Based on these values, it can be concluded that the person has the disease SNP. Since these results Error that may have resulted during the PCR & detection steps could be that the samples were exposed to too much light. If the samples were exposed to more light than the localized beam, such as improper filtering of light with the box, it would cause less fluorescence to be emitted from the samples because it would have been exposed to a frequency other than the laser beam. Another could be incorrect analysis of the samples using ImageJ. The ImageJ analysis could cause error by not analyzing the sample accurately, such as when highlighting the ellipse around the sample, or by having too much light surrounding the sample. Another possible source of error could be an inadequate photo quality. For Calculations 3 and 4, the Bayes values are very small, indicating that the PCR is not very reliable for predicting the development of the disease, because the person will most likely not get the disease. Computer-Aided Design TinkerCAD TinkerCAD is a 3D computer-aided design (CAD) tool, which operates completely on the Internet. Using a variety of tools, such as highly adjustable shapes and symbols, and grouping and color tools, 3D models can be designed on TinkerCAD and actually produced as objects through 3D printing. To design the OpenPCR machine, various STL (STereoLithography) files representing different parts of the machine were imported from Thingiverse, a website used to share digital design files, to TinkerCAD. The tools on TinkerCAD were then utilized to create a newly designed OpenPCR machine that addressed the limitations that existed with the original design. The components of the consumables kit and fluorimeter were designed entirely on TinkerCAD without the use of imported STL files from Thingiverse. Our Design Feature 1: Consumables Kit For packaging the consumables, we will place the primers in a more convenient bottle, perhaps in easy, disposable containers. For the problems within the consumables, it is a hassle to consistently the PCR mixes. There will be a method to simply make it easier for transferring the PCR mix and primers into the DNA samples so that it doesn't leave room for error. Feature 2: Hardware - PCR Machine & Fluorimeter Fluorimeter For the Fluorometer, we wanted to get rid of several sources of error, so we wanted to have a Fluorimeter with a built in camera and touchscreen interface as well as the use of six-hole slides. The built in camera allows the machine to analyze the drops on its own and eliminates the need for a phone cradle which also eliminates a lot of error due to how awkward the phone cradle was and how it caused the fluorimeter setup to be in and out of light constantly. The touch screen interface increases utility as it is easy to use. Finally, the six-hole slides will allow for the camera to analyze the drops from below as well as ensure the drops form properly and reduces the amount of waste. (insert image) an auto-recorder for the data. For the OpenPCR Machine, we decided that a larger heating block was needed to decrease the amount of time for testing. We also wanted to implement more fans to speed up the thermal cycling process for the PCR to occur. (insert image) OpenPCR Machine Back OpenPCR Machine Side OpenPCR Machine Top An issue within the OpenPCR Machine is that the machine is slow (it took a couple of hours and required the team to leave the samples in other’s care). And finally, the issue within the Fluorometer System is that it had to be completely dark when the picture was taken, and an iPhone timer had to be used, which added some hassle and room for error (if it took the picture before it was fully dark).