BME100 f2016:Group4 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 Error creating thumbnail: Unable to save thumbnail to destination Jaeger Moore Error creating thumbnail: Unable to save thumbnail to destination Name: Mahina Wing Error creating thumbnail: Unable to save thumbnail to destination Name: Matt Morton Error creating thumbnail: Unable to save thumbnail to destination Name: Trevor Wood Our Brand Name LAB 6 WRITE-UP Bayesian Statistics Overview of the Original Diagnosis System In the BME100 10:30 class, 17 teams of 4-6 students diagnosed 26 patients total. In our small group of 4 people, we divided into 2 teams of 2. Each team divided the lab to increase efficiency. For example, one team completed the PCR Reaction and the other group conducted research on the human genome. Our group was given 2 patients' SNP. To prevent error, we used 3 replicates for each patient. Every group in the class used the following specific PCR controls: HEATED LID: 100°C INITIAL STEP: 95°C for 2 minutes NUMBER OF CYCLES: 25 Denature at 95°C for 30 seconds, Anneal at 57°C for 30 seconds, and Extend at 72°C for 30 seconds FINAL STEP: 72°C for 2 minutes FINAL HOLD: 4°C We also used took 3 images of each specific PCR sample to compare to each other and get better data. On ImageJ we had to make sure we measured the RAWINTDEN came from the same amount of area from each picture. To get the best RAWITNDEN we also tried to keep the area in relatively the same area for each picture, avoiding bright spots. One major challenge our group had was working with Image J. Originally we worked on 2 separate computers. This means that the area used to measure RAWINTDEN was different for the different images. During our calculations, we quickly found out that this badly affected the accuracy of our results. We then had to redo all the ImageJ measurements with the same area. What Bayes Statistics Imply about This Diagnostic Approach The results from the calculations (1&2) suggest that individual PCR replicates are usually quite accurate in determining the final PCR conclusion of a sample. When replicates tested positive for the disease SNP, the final conclusion also turned out to be positive almost 90% of the time. While not quite as accurate as one might expect, 9 out of 10 samples is fairly good. When an individual PCR replicate tested negative for the disease SNP, the final PCR conclusion turned out to be negative almost 95% of the time. This is very good, and it means only 5 out of every hundred samples will test negative when the PCR conclusion turns out to be positive. If three samples are used for each final conclusion, two out of three would be negative only 0.25% of the time when the actual result was positive. The results from the calculations (3&4) suggest that the PCR test is very unreliable for detecting whether someone actually has the disease, and is somewhat unreliable for detecting if someone does not have the disease. When a positive final test conclusion occurred, the patient turned out to have the disease slightly less than 50% of the time. This means that the majority of the time, when the patient tested positive, they actually did NOT have the disease. This is extremely troubling for a diagnostic test. On the other hand, when a negative final test conclusion occurred, the patient turned out to be disease-free 4/5's of the time. While this is much closer to 100%, for every 5 patients who test negative, one actually does have the disease. In a situation where treatment was necessary, this would not be acceptable. Overall, in terms of the results from the class, this PCR test is not particularly reliable whatsoever. Sources of Error The first source of error could have been during pippetting for the PCR reaction. It is possible that we did not use exactly the correct volumes to mix in each tube. This could have effected our results. For example, if we did not use the correct amount of Cyber Green, then the green light in our images could be more or less than it should have been. This would have effected our ImageJ results and ultimately effecting our Bayes Analysis. The second source of error could have been in our ImageJ measurements. As mentioned earlier, we originally used different areas to measure the RAWINTDEN. While we redid the measurements to fix the issue, the location of each area on the water drop varied. The different locations could have had more concentrations of light effecting our data for the Bayes Analysis. The third source of error could have been in our calculations of the constants for the Bayes Analysis. The constants were supposed to be based off the conclusions from each groups' results. However, some groups did not turn in their data and had no conclusions. This gave us less data to consider into our constant which could have effected its accuracy. These somewhat inaccurate constants would negatively affect our Bayes analysis. Intro to Computer-Aided Design 3D Modeling Our group used Solidworks to create a structural improvement to the flourometer machine. To start out, we had to make a rectangular prism with the dimensions of the flourometer machine. This was simple using the sketch and extrude emboss. Additionally, the cutout where the droplet plate goes needed to be created with the extrude cut tool. Another extrude cut needed to be made for the new sliding holder to insert into. Since the parts were all fairly uniform and we only had to use three or four tools in Solidworks it was fairly simple. Once both our parts were done (the sliding phone holder, and the flourometer machine), we had to create an assembly. When doing this, we were able to allow the sliding part to slide back and forth to mimic the device. Our Design In order to make our product user friendly we decided to improve the flourmeter and make it easier to record the presence of DNA. We designed a device that attached to the bottom of our flourmeter allowing the user to slide their phone up to the face of the box as well as away and in the vertical position of the phone, as well as having a built in ruler to be able to note the exact distance. This not only improves the ease aspect but also removes possibility of error through light being able to get in as well as human error of position and holding the phone. Feature 1: Consumables In our design we have decided to keep all the original consumables. This includes the liquid reagents: PCR mix, primer solution, SYBR Green solution, and buffer. We will also include the plastic tubes, glass slides and pipette tips. However, our kit will present all these consumables in a user-friendly, easy to understand packaging. We will arrange each of the consumables in order of steps needed to complete the PCR. Preparing the PCR samples was a hassle. We had to put on the pipette tip,transfer PCR mix, discard the pipette tip, get a new pipette tip, and then transfer the DNA to the same tube. All these steps were on different trays and spread out around our table. With our design, all the steps are located in order on one tray. For example, the tray would have a row of tubes in this order: a pipette tip, PCR mix, pipette tip, DNA sample. Underneath this row would be the empty tube the mix is going into. This way the user simply goes down the row of tubes in order placing into the empty tube. We would also make a similar tray that does this for when we had to add SYBR green to the tubes. This makes it easier and more efficient to perform the PCR. Feature 2: Hardware - PCR Machine & Fluorimeter In our system of running PCR we have decided to keep the basis of our process of running the PCR the same in using the thermocycler the exact same. There is no outstanding weakness to the thermocycler and we see far too many strengths to the machine to make any significant changes to it. The machine that we decided to use in our process is the same style of thermocycler that we used in class It's lightweight which is a huge bonus in lab as many older PCR machines that you find in other labs can be twice to five times the size. In more devolved countries mobility may be a non-factor, bit in countries where labs and medical clinics are constantly on the move this would be a huge bonus to whoever may be running a PCR reaction to identify genetic and transgenetic diseases that may be found in a patient. Keeping it lightweight makes it easier to be mobile and simple to set up lab anywhere so that a scientist or whoever may be running the PCR can focus on the accuracy of their reaction and analysis of data to help the people they are running the PCR for. The thermocycler uses an Arduino microcontoller that is programmed on how the whole machine carries out the process which is a huge bonus. With a microcontroller such as the Arduino Uno being in our machine, the thermocycler can be reprogrammed in any way that a person running any PCR reaction can control the temperature changes that occur during a given PCR process. This is incredibly important in running PCR due to the fact that there are selective portions of DNA that denature at certain temperatures and when somebody is trying to isolate/amplify a very spesific portion of DNA the ability to change the max temperature of a cycle ran in the machine is critical to amplifying the portion of DNA being studied by whoever may be running the reaction. The use of a micro-controller also minimizes the amount of human error that could occur if the heating and cooing of the PCR were carried out by hand. The cooling system of the thermocycler is also fairly efficient even if the fan and the heat sink cooling system takes up a large part of the interior of the thermocycler, it is not necessary to change the design due to the ease of manipulation of the process due to the reprogram-ability of the arduino being such a large plus factor to the universal use of the thermocycler. The cooling system easily has the capability to cool items to the temperature of 160°C which at that point any human DNA being ran in PCR would definitely denatured enough to be subject to study. The fluorimeter aspect of our system is something that we have decided to change due to some weaknesses that we experienced in our uses of Image J, phone cameras and keeping a uniform distance from the liquid sample we were capturing. The problem we experienced with image J and the phone camera is that using Image J was difficult as the phone camera that we used had an obscene amount of glare due to the blue light that we used on the florometer stand. image J was just a weird and really time consuming process to use just to find the concentrations within our sample and keeping the area of the circle on the program was also a problem and was just not user friendly at all. To solve this we have decided that creating a phone app version of image J would be the best option to minimize time to analyze the concentration of samples. The way the app would work is that you open the app on your phone and position it on out stand (explained later in this paragraph) so that it is at the optimal position to record the data wanted and then click a button that will read "capture" and then the image J app will carry our all the processes we did in a previous lab to find the concentrations in our samples. We could program the app to carry out all necessary calculations to determine concentrations of certain sample which would minimize the amount of human error that could occur in any sort of analysis of sample. The advantage of this app is that the data that is needed out of the use of this image J app could be displayed on the phone screen giving the user immediate feedback on whatever data is needed for analysis of the sample. Conversions and such can also then be done on the phone itself rather than carried out by the user or a calculator, minimizing even more error and making things much more simple to use. This app will also compensate for the glare by editing out the glare found on the input images used for analysis on the image J app. This sort of retouching on photos is automatically done on applications like Photoshop so we could code into the app to carryout that process while it is capturing images. The combo of these two processes in our app would make the process of finding concentrations of samples using photography simple easy and very user friendly making it simple to analyze PCR reacted samples. The third major weaknesses for the Flourimeter machine was that there was no way to maintain a uniform distance between the phone holder and the sample drop. Although we were able to measure, small shifts in the components as the shade was put in place changed this distance and created blur/inconsistency. Another problem was how the camera did not line up with the sample drop, so stands had to be placed under the Flourimeter (not very convenient). To solve both these problems, we created a sliding connector between the machine itself, and the phone holder. The sliding component has the ability to slide on tracks to adjust the height of the holder to align with the drop, and another set of tracks to adjust the distance from the drop. The distance would be marked with measurements on the sliding component, and be able to lock in place to maintain the same distance on each picture. Our changes to the fluorimiter portion of analyzing our samples will make anybody who adopts our technology and technique.