# BME100 s2015:Group3 9amL6

BME 100 Spring 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 TEAM

 Emily Santora Christina Salas Steven Mills Austyn Howard

# LAB 6 WRITE-UP

## Bayesian Statistics

Overview of the Original Diagnosis System

Eight groups of approximately four students each diagnosed 16 patients total. Everyone did an equal amount of work. Specifically, for our group, two people were in charge of the Excel document and calculating results. Also, one different person was in charge of checking each result on his or her calculator. Lastly, the last person recorded the data onto a piece of paper and presented them in an easy-to-read manner. To prevent error, we calculated the values for each variable using both Excel and a calculator. Furthermore, we used controls to ensure the solutions being used to test the patient samples would provide accurate results. Group five's results were disregarded due to inconclusive data. There was one other inconclusive conclusion, eight negative conclusions, and five positive conclusions. The class results are provided below, and the Bayesian statistics were calculated using these results:

Class Results:

Plugging in values collected to P(A|B) = (P(B|A) * P(A))/P(B)

Calculation 1:

P(A) Total positive conclusion frequency = (0.357143)

P(B) Total positive frequency = (0.380952381)

P(B|A) Total positive frequency given positive conclusion = (0.875)

P(A|B) = 82% --> given a positive PCR reaction, there is a 82% chance of a positive Diabetes conclusion

Calculation 2:

P(A) Total negative conclusion frequency = (0.571429)

P(B) Total negative frequency = (0.595238095)

P(B|A) Total negative frequency given negative conclusion = (0.88)

P(A|B) = 84% --> given a negative PCR reaction, there is a 84% chance of a negative Diabetes conclusion

Calculation 3:

P(A) "Yes" diagnostic for Diabetes = (0.4285714)

P(B) Positive PCR conclusion frequency = (0.357143)

P(B|A) "Yes" diagnostic with positive PCR = (0.6667)

P(A|B) = 80% --> given a positive DNA test conclusion, there is a 80% chance of a "yes" diagnosis for diabetes

Calculation 4:

P(A) "No" diagnostic for cancer frequency = (0.5714286)

P(B) Negative PCR conclusion frequency = (0.571429)

P(B|A) "No" diagnostic with negative PCR = (0.75)

P(A|B) = 75% --> given a negative DNA test conclusion, there is a 75% chance of a "no" diagnosis for diabetes

Calculations 1 and 2:

First, calculation 1 implies that the reliability of the individual PCR replicates for concluding that a person has the disease SNP or not is high that, given a positive PCR reaction, there is exists a positive diabetes conclusion, because it is close to 1.00. Further, calculation 2 implies that the reliability of the individual PCR replicates for concluding that a person has the disease SNP or not is high that, given a negative PCR reaction, there exists a negative diabetes conclusion, because it is close to 1.00. There are a few possible sources of human or machine/device error that could have occurred during the PCR and detection steps that could have affected the Bayesian values in a negative way. First, contamination could have occurred during the experiment, which would have caused some results to show up as positive, negative, or inconclusive when they actually were supposed to show a different conclusion. Also, some groups might have taken their pictures of their drops at different lengths from the drops, which would have caused errors. Lastly, specifically, our group had trouble using ImageJ; we could have used the software wrong and obtained faulty data.

Calculations 3 and 4:

First, calculation 3 implies that the reliability of the PCR for predicting the development disease is high, given a positive DNA test conclusion, there is exists a "yes" diagnosis for diabetes, because it is close to 1.00. Further, calculation 4 implies that the reliability of the PCR for predicting the development disease is high, given a negative DNA test conclusion, there is exists a "no" diagnosis for diabetes, because it is close to 1.00.

## Computer-Aided Design

TinkerCAD is a 3D modeling online program that helps people design 3D objects with little to no experience. It allows people to create specific parts with 100% customize-ability, then you can group objects together to create a single machine with everything scaled to proportion. TinkerCAD was used by our group to recreate the PCR macheine using templates provided by Dr. Haynes. TinkerCAD is a very useful tool for the user to visualize how a product will appear.

Our Design

This image shows the new and improved Flourimeter that was created using TinkerCAD Computer Aided Design. The design is a single box that has the camera, and the light emitter inside of it, and two doors with the elbow hinges on the outside, so the doors sit flush with the box eliminating most if not all of the excess light. The doors will have handles on them to easily open and close them.

## Feature 1: Consumables Kit

The consumables kit provided for the PCR machine will include:

• Liquid reagents - PCR mix and Primers
• Pipettor
• Pipette tips

Each item will be packaged individually to ensure no leakage of the liquid reagents or damage to the pipette equipment. The liquid reagents will be sealed in PCR recyclable containers made from HDPE Plastic. This plastic is made from 100% recycled plastic bottles and performs as virgin plastic. The caps on the bottles will have a rubber twist on cap to ensure a tight seal to avoid leakage from the liquid reagents. The containers will then be placed in specially designed bags that filter out light that will damage the integrity of the reagents. Once the viles are completely sealed, they will be placed in a separate compartment of the package enclosed with movement absorbing packing peanuts. The pipettor will be placed in a Styrofoam mold preventing damage during transportation. The sterilized pipettor tips will be placed in sealed package and also placed a Styrofoam mold to prevent unnecessary movement during transit.

The packaging for the consumables kit is specifically designed to counteract any drops, falls, crushing and any other unforeseen maltreatment of the parcel during transit. The Styrofoam both in solid and peanut form will protect each component of the consumables kit from unnecessary harsh movement. The caps on the liquid reagents are an extra precaution during travel and once the reagents are in use in the lab. The caps, when properly tightened on the containers, will prevent leaks and spills. The sterilized bag for the pipettor tips will prevent accidental contamination of the tips during transit and in the lab until the tips are needed. Once the bag is open, the tips are secure in a plastic box that will avoid contamination from spills inside the lab.

## Feature 2: Hardware - PCR Machine & Fluorimeter

Our system will contain the PCR machine, unaltered, with a newly designed fluorimeter both connected to a computer for testing. Samples will be amplified in the PCR machine and then transferred to the flouimeter for imaging. A built in camera will take pictures of the samples by computer access and will automatically analyse the samples.

A major weakness that our group found out about the fluorimeter was the clunky and non-user friendly design. Also the current design let a lot of light in, possibly interfering with the picture quality and leading to false data when analyzing the pictures. In order to fix these problems we designed a new and improved fluorimeter with hinges and door handles to create a closed box when taking pictures, and a less clunky design. The design is a single box that has the camera, and the light emitter inside of it, and two doors with the elbow hinges on the outside, so the doors sit flush with the box eliminating most if not all of the excess light. The doors will have handles on them to easily open and close them.