BME100 f2015:Group8 1030amL6

BME 100 Fall 2015	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: Sydney Connor Name: Michael Dagher Name: Tajinder Virdee Name: Angela Hemesath Name: Olivia Gonzalez Name: Alaina Jenish LAB 6 WRITE-UP Bayesian Statistics Overview of the Original Diagnosis System 17 teams of 6 students diagnosed 34 patients for a disease caused by a single nucleotide polymorphism (SNP). 2 patients were tested for each team using PCR to detect the SNP. In order to limit error, 3 samples were taken for each patient so that the chances of having a false positive or negative result would be reduced, producing a total of 6 samples per group and 102 samples total.In addition, previous to testing the PCR samples, a calibration curve was set up using ImageJ for samples with known DNA concentrations. This allowed a basis for comparison for the PCR samples of the patients, whose DNA concentrations were unknown. Moreover, each team used a positive and negative control in tandem with testing the patient samples during the PCR, in order to verify that their methods were correct and that the PCR had been done properly. The drop image of each sample during testing for fluorescent levels was taken 3 times, in order to reduce error by allowing any variations to be taken into account. The final results included 32 patients with conclusions, and 2 patients with blank results. Of the successful conclusions, 13 were positive, 17 were negative, 2 inconclusive results, and 2 "not test" results. For this group, specifically, challenges during the process included a skewed calibration curve, in which a zero concentration of DNA did not correlate to zero detected fluorescence. This resulted in negative concentration values for the actual PCR samples. However, these samples' concentrations were still lower than that of the negative control, so the group was able to maintain that the patients were negative fro the disease. What Bayes Statistics Imply about This Diagnostic Approach In calculation 1, the probability of positive a final test conclusion was around 50%. Similarly, the probability of a positive PCR result was around 50%. This meant that the probability that a patient would get a positive PCR reaction, given a positive result was around 80%. Similarly, the probability that a patient would get a positive result, given a positive PCR reaction was also around 80%. These last two probabilities are relatively high, indicating that a positive PCR result is mostly reliable in predicting a positive disease conclusion. In calculation 2, the probability of a negative final test conclusion was around 50%. Similarly, the probability of a negative PCR result was around 50%. This meant that the probability that a patient would get a negative PCR reaction, given a negative result was around 80%. Similarly, the probability that a patient would get a negative result, given a negative PCR reaction was around 90%. These last two probabilities are relatively high, with the final probability even higher than the previous one as well as the final probability in calculation 1. This indicates that a negative PCR result is very reliable in predicting a positive disease conclusion, even more so than a positive PCR result is in predicting a positive disease conclusion. Thus, both the positive and negative PCR tests are very reliable in predicting a positive and negative disease outcome, respectively. However, a negative PCR test is more reliable than a positive one. In calculation 3, the probability that a patient would develop the disease was around 30%. Like in calculation 1, the probability that a person had a positive final test conclusion was around 50%. This meant that the probability that the patient would get a positive final test conclusion, given that the patient developed the disease was around 30%. And, so, the probability that the patient would develop the disease, given a positive final test conclusion was around 25%. These final probabilities are relatively low (less than 50%). This indicates that the PCR final test conclusions are very unreliable in predicting a positive diagnosis of the disease. In calculation4, the probability that a patient would not develop the disease was around 70%. As in calculation 2, the probability that the patient received a negative final test conclusion was around 50%. This meant that the probability that a patient would get a negative final test conclusion, given that they do not develop the disease was around 70%. The probability that a patient would not develop the disease, given a negative final test conclusion was around 90%. These final two probabilities are relatively high (above 50%), especially the final probability. This indicates that a negative PCR test conclusion is very reliable in predicting no development of the disease. Thus, it can be concluded that a positive PCR final conclusion is very unreliable in diagnosing the disease, with a probability of being correct at around 25%. This means that less than half of the patients who received a positive PCR final conclusion would actually develop the disease. However, a negative PCR final conclusion was very reliable in diagnosing the disease, with a probability of being correct at around 90%. This means that nearly all patients who received a negative PCR final conclusion would not develop the disease. Human or machine error may have affected the Bayes values mentioned above. These errors may include an incorrect amount of PCR mix or patient sample used during the PCR process due to an incorrect handling of micropipettes, an skewed fluorescent reading of the samples, due to differences in phone cameras, and a failure to create a "beach ball" shape during the detection ster. In the first source of error could have greatly affected results and their ability to predict the disease. If not enough of the PCR mix or the samples were used, the PCR may not have worked correctly, giving incorrect data. Alternatively, if different amounts for each sample were used, the PCR would work for some samples but not others, giving unreliable data and skewing the calculated statistics. Alternatively, each group used different cameras with different settings, depending on the camera's capabilities. This means that results read by one group may have differed from another's, so that the same data may have different results. This may have also skewed the calculated statistics. Finally, if the drop was not placed correctly during the detection step, so that the oval shape of the drop was breached, this may have made a reading of the fluorescence different, since any imperfection would reflect the light differently, causing a higher or lower reading than what is actually the case. This may have also skewed the data, and thus the statistics as well. Intro to Computer-Aided Design TinkerCAD The TinkerCAD tool is a more visual design software, shapes are pre-made for the user. Therefore, the user just has to click on the shape and adjust different aspects of the design to make a product. TinkerCAD can be compared to SolidWorks because SolidWorks is a lot more mathematical and uses more precise measurements. SolidWorks is also more precise with regards to the placement and assembly of the different parts of a product. TinkerCAD's zooming tool is also not very fine tuned, whereas SolidWorks is very exact whenever you zoom in or out of the object of interest. TinkerCAD is essentially easier to use and more user friendly whereas SolidWorks requires a lot more attention to detail in order to make a product. Essentially, SolidWorks is the Gold Standard of design software whereas TinkerCAD is a more accessible and user friendly version of the SolidWorks that engineers work with. Our Design This design features solar panels (shown in red) that power the device. The green area is also used as a fluorometer so its essentially an "all-in-one" device. The design also features 40 PCR-tube holders to allow for the maximum amount of samples in a condensed space. There is also the ability to hold different sized samples. This particular design was chosen to allow the user to save time and money by just buying one device instead of two. Since the device is also solar powered, it is environmentally friendly and would allow it to to be used in countries where electricity inst as stable. Feature 1: Consumables When a buyer purchases one of the reputable Sun PCR machine, they can use it anywhere under the sun. This portable PCR includes PCR reaction mix, SYBR Green solution, and a buffer solution. The customer can use any any type of PCR tube, the Sun PCR can accommodate for the .5 mL as well as the 5 mL PCR test tubes. The original PCR included PCR Reaction mix, PCR tubes, DNA and Primer solutions, and pipette tips. The Sun PCR eliminated the pipette tips and the PCR tubes because the machine itself accommodates for the standard sizes of each, so the user can buy any pipette tips or PCR tubes and they should be able to use them in this PCR machine. The Sun PCR kit also does not include any DNA or Primer solution, as the Sun PCR machine is not made to test one specific disease, but rather it is able to accommodate for different types of tests and therefore the DNA and primer solution will vary depending on the user's intention with the Sun PCR itself. The original design included the appropriate consumables to test one specific disease, however it was to specific and those specific consumables could not be used for other tests, so Sun PCR will be able to accommodate for different types of disease testing.Therefore since the Sun PCR machine is such a multipurpose PCR machine, the different types of DNA and Primer solutions would be too numerous and expensive to formulate a kit to distribute to all buyers of the product. Therefore the user of the Sun PCR machine will have to purchase DNA and Primer solutions from an outside source in order to do a specific test that they have in mind. Feature 2: Hardware - PCR Machine & Fluorimeter The Open PCR machine included in the original lab had the benefits of being lightweight, cheap, easy to construct, easy to use, and easy to get data from. However, it was limited in how many PCR samples it could process at once, could only process PCR samples in a specific tube size, and consumed a lot of energy. To rectify this, though we kept the general structure of the PCR machine to maintain the lightweight, cheap, easy to construct, easy to use, and easy to obtain data from aspects, instead of using a plug for power source, we would instead use several small solar panels, and would expand the heating block to include 32 standard small PCR tubes, and 8 standard large PCR tubes, to offer variety. The PCR system was made cylindrical, to minimize surface area and material cost, as well as minimize the number of parts to put together, so that construction would be easier. The USB to access data from the PCR machine was maintained. The PCR machine was made out of the same cardboard materials, in order to minimize cost, and to make the machine lightweight. The benefit of the fluorimeter was that it had few components and was easy to construct. However, it was bulky, difficult to use, slow to process, required outside materials, which varied from person to person, and was inexact. To maintain the easy construction of the fluorimeter, the same materials were used for the outside part, and the number of separate parts to put together was minimized, and made easy to understand. However, to rectify the downsides of the fluorimeter, many changes were made. The fluorimeter was attached to the bottom of the PCR machine, making the entire system easy to do in one sitting, and limiting the bulkiness of the fluorimeter. Moreover, there was a lot of difficulty and time wasted during the setting up of the phone and then finding the correct place to put the phone for it to focus. So, instead of using a phone at all, the fluorimeter will have a laser and sensor installed, so that the distance and the readings are the same for every sample and every individual. Moreover, to avoid any shaking, or disturbance of the samples, and to maintain total darkness, the fluorimeter will include multiple drawers with places to put the sample drop on. These drawers will easily slide into the box, and the samples, placed in a row, will be analyzed equally the laser passing through to the sensor, while in total darkness. This limits error, and allows for multiple readings at once. The fluorimeter would also include the USB port that the PCR machine had, in order to continue the easiness of data access. This way, the data can simply upload from the machine to a computer.