Overview of the Original Diagnosis System
BME100 successfully tested and diagnosed 34 patients for the disease-associated SNP. Teams of 6 individuals were responsible for 2 patients each – a combined effort of 17 teams. This allowed each team to wholly focus on only 2 patients rather than sacrifice the quality of the diagnoses in order to test more patients. Several methods were exercised to prevent error; these methods included the number of replicates per patient, the replication and testing of pure PCR positive and negative controls, and various other methods. During testing, micropipette tips were changed after every usage, glass slides were changed between each sample of PCR product and SYBR Green I, and thorough labelling of test tubes and materials were done to eliminate the risk of cross-contamination between samples. The Java program used, ImageJ, had specific calibration controls of its own that minimized error. Aside from taking three images of each PCR product and SYBR Green I solution, the RAWINTDEN of both the drop and the background were measured and subtracted – eliminating any discrepancies in light intensity within the box where the fluorimeter was located. The distance between the phone camera and the PCR product - SYBR Green I solution was initially measured and kept the same throughout the lab as another calibration control.
In total, there were 30 successful tests, 2 inconclusive tests, and 2 blank tests. Of the 30 successful tests, 13 tested positive for the disease-associated SNP and 17 tested negative for the disease-associated SNP.
During the lab, one of the main challenges came with successfully using the fluorimeter. The smartphone stand could easily be bumped – either messing up the distance accuracy between the smartphone and the fluorimeter or knocking over/jostling the smartphone itself. In addition, the absence of 2 results and the 2 inconclusive results could skew the frequency calculations.
What Bayes Statistics Imply about This Diagnostic Approach
Calculation 1: What is the probability that a patient will get a positive final test conclusion, given a positive PCR reaction?
Calculation 2: What is the probability that a patient will get a negative final test conclusion, given a negative diagnostic signal?
Calculation 3: What is the probability that a patient will develop the disease, given a positive final test conclusion?
Calculation 4: What is the probability that a patient will not develop the disease, given a negative final test conclusion?
Calculation 1 shows that about 75% of patients that have a positive test result had a positive PCR reaction. THis means that 25% of people who have a positive test result could have a negative PCR reaction. The second calculation states that about 90% of people who have a negative final test result will have a negative PCR result this means that about 10% will have a negative final test result but a positive PCR result. Both calculations do not have to do with patients actually having the probability of getting the disease so patients should not be worried about these type of calculations.
According to calculation 30% of the patients will develop the disease. Despite this, 40% of the patients tested would receive positive test results. This means that the probability of a patient getting a positive test result and actually developing the disease was 30%. This inaccuracy means that far more than half of the patients would be likely to receive false positive test results leading to unnecessary panic. According to question 4, there was a 70% chance that a patient in the group would not develop the disease. However, there was a 50% chance that a patient would receive a negative test result. Consequently, the likelihood of a patient receiving a negative test result and not developing the disease was only 70%. While the negative results are more reliable than the positive results, there is still a large potential for error that would lead to many patients receiving false negative results that would then go untreated.
One possible source of error in during the PCR lab may have been human errors in measuring. While micropipettes are the most accurate way to measure such small volumes it is possible that measurements were slightly off. Another source of error may have been inaccurate temperature control during the PCR due to the machine itself. While the machines used were high tech pieces of equipment they were not the “best” on the market today. While it is not absolute it is possible that these machines affected the final Bayes values. Finally, an additional and more probable source of error may have been the inaccurate results during the fluorescence detection step with the smartphone. Due to the insufficient supplies available, use error may have occurred due to excessive exposure to light or inaccurate pictures that caused errors in the data collected from Image J.
Intro to Computer-Aided Design
TinkerCAD
TinkerCAD was a free design tool that allowed us to make shapes and fit them together in the shape of a Open PCR machine. We could also change the color of certain parts to go along with the design colors. Also if we wanted to we could put our motto and team logo on the machine.
Our Design
This design solely improves upon the efficiency, transportability, and tidiness of the OpenPCR design. Rather than run the risk of misplacing or forgetting a necessary PCR component, a storage compartment on the side of the OpenPCR machine has the ability to hold a micro-pipette, disposable micro-pipette tips, gloves, and PCR tubes. In addition, the shell/hull of the compartment is a waterproof insulator - keeping the temperature inside the compartment relatively constant and dry, and preventing leakages should anything open up and spill during travel. While all trays within the compartment have securable lids, one 8 tube tray has a black out lid - preventing any additional light from hitting tubes filled with light-sensitive material while the compartment itself is open.
Label 'A' on the 2nd image is the micro-pipette holder. The top two bands are elastic and allow for easy placement and removal while the bottom funnel ensures that the pipette does not move during transport and remains upright where it runs the least risk of affecting its calibration.
Label 'B' denotes the PCR tube trays, all covered with securable and transparent lids.
Label 'C' denotes a partially solid and partially mesh compartment for several pairs of rubber gloves.
Label 'D' denotes the micro-pipette tip trays, also covered with a securable and transparent lid.
Label 'E' denotes the 8 tube tray with a non-transparent lid that can be slid up and down for easy access to light-sensitive solutions.
Feature 1: Consumables
This consumables kit includes only the consumables necessary for basic PCR, not any processes occurring post-PCR on the PCR products.
Included Consumables:
Micro-Pipette
Full tray of disposable micro-pipette tips
Full tray of PCR tubes
1 set of small, medium, and large gloves
This consumables packaging plan addresses the major weakness of disorganization during preparation for PCR and even any processes occurring after PCR on its products. Rather than managing all necessary components separately and risk forgetting or misplacing a component, everything necessary to successfully operate and utilize OpenPCR is all in one place - with the OpenPCR machine. The bench will always be neat, tidy, and compact, only requiring an additional waste bin for disposing of the gloves and pipette tips.
Feature 2: Hardware - PCR Machine & Fluorimeter
Only the OpenPCR machine with the redesigned and implemented storage compartment will be included in the system. The fluorimeter and any other hardware and materials that are relevant to PCR are not included. The system focuses solely on making the OpenPCR system more efficient and organized. It would be improbable and unrealistic to also account for any and all hardware used on the PCR products when redesigning the OpenPCR system for efficiency.
BME 100 Fall 2015
