BME100 f2014:Group30 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 Name: Megan Wieser Name: student Name: student Name: student Name: student Name: student LAB 6 WRITE-UP Bayesian Statistics Overview of the Original Diagnosis System In the PCR lab experiment, the DNA of 68 patients total was amplified and analyzed for a specific disease indicator in order to predict the group's probability of disease detection. To divide the work for this experiment, 34 groups of 6 people each tested 2 patients to determine the likelihood that either patient would contract the disease. As with any experiment, steps were taken to prevent error. For example, 3 total replicates were tested for each patient as a way to minimize the likelihood that the patient will be misdiagnosed. In addition, the PCR mix had positive and negative control solutions, which helped to make sure that the PCR machines and fluorimeters produced the expected results for the known positive and negative solutions. Regarding the Image J processing, calibration controls of positive and negative solutions were utilized to prevent error in the sense that they provided a baseline for what was supposed to happen in patient samples with negative and positive test results. In other words, the positive control showed the SYBR green, which indicated that the PCR and fluorimeter machines worked to produce expected, controlled results, and it also indicates that other patient samples with the same green color test positive. To increase the accuracy of the Image J processing values, 3 drop images were taken per PCR sample and averaged together. Although the class's final data contained a majority of successful conclusions, there were discrepancies within the data collection that affected the overall Bayesian calculations for the specificity and sensitivity that a patient has a disease. For example, 8 out of the 68 patient results were inconclusive, which means that the calculated Bayesian probability statistics were higher or lower than the actual probability depending on whether the inconclusive results should have been positive or negative. Blank data was also discrepancy within the data collection, as the 6 blank conclusions decreased the total PCR conclusions that were used to calculate Bayesian probabilities. Since the total number of conclusions was decreased due to the number of blank conclusions, the statistics were higher than what they should have been if there had been 68 total conclusions. For example, if 20 out 68 people had a positive conclusion, there would be a higher probability of a positive test conclusion than if 20 out of 62 people had a positive test conclusion. Measures were taken to obtain the most accurate data collection, but there are things like inconclusive and blank test results that can still affect the data analysis. What Bayes Statistics Imply about This Diagnostic Approach Computer-Aided Design TinkerCAD Our Design Feature 1: Consumables Kit Feature 2: Hardware - PCR Machine & Fluorimeter