Weley Lin, Paul Ngo, Tu Truong, Jayna Wittenbrink
Sponsor: Dr. Sandrine Miller-Montgomery
Each of our bodies has a microbiome made up of trillions of microbes and their genes that affect our health. Microbiomes also exist all over the Earth in animals, plants, and bodies of water, which serves as an indicator of the health of the planet. The microbiomes that live in our gut help us digest and process nutrients, interact with and shape our immune system, and are associated with diseases--some that are unexpected including rheumatoid arthritis and asthma. To understand more about the microbiome in our body, many more human samples are required to be analyzed. The Center for Microbiome Innovation at UC San Diego focuses on the immense data generation needed to move the results from research to improving therapies and advancing applications in the field.
The benefits of this method is that the tube with cotton-swab assembly is cheap (~$1.50 per set) and lab processing is highly efficient (throughput of 100k samples/year). However, the cons with this is that delivery of the sample is more difficult. During transport, the tube is subject to damage and the environment can promote unwanted bacterial growth.
The method that our sponsor would like to transition to is using sampling cards such as the FTA or FOBT card for sample collection.
FTA Card FOBT Card
However, the major issue with using sampling cards is the risk of cross-contamination. Using a physical tool on the market, such as a hole puncher, results in direct contact with the cutting mechanism and DNA material. This leads to cross-contamination issues if another sample needs to be cut using the same tool. Current methods of decontamination such as using an autoclave, bleach solution, or UV rays is a time-consuming process, which significantly decreases throughput. The estimated throughput with this method is currently about 10k samples/year (10% of the throughput using tube/cotton-swab).
The basic requirement for our project is that a device is created to cut out samples, or “chads”, out of the FTA sample card to fit into the wells on the 96-well plate without cross contamination.
Final Design Overview
Our final design uses a single mechanical punch die system to cut out sample chads from FTA sample cards. The decontamination method uses in-between clean paper punches. The 96-well plate sits on top of a holder that is attached to an XY stage, which is used to position each well with the punch hole.
The machine has 4 main assemblies:
1. Cutting mechanism
2. Clean paper feeder
3. XY stage + 96-well plate holder
4. Machine housing
The cutting mechanism is a single punch die system that uses a punch head attached to a linear actuator to operate. When activated, the punch head shears the FTA card or clean paper against the die hole located at the center of the die plate. This shearing force cuts the paper into a circular shape called a chad.
To decontaminate the punch head to prevent cross-contamination, the punch head cuts clean paper twice following a sample card cut. Analysis was conducted by the Knight Lab that suggests this is a sufficient decontamination method. The clean paper comes from a paper roll attached to the side of the housing and is fed over the punch area by a geared motor.
All chads fall into a well of the 96-well plate, which is aligned with the die hole. The 96-well plate is accurately positioned by the XY table assembly below. This increases throughput by eliminating a step that the lab technician would have to manually perform.
All of these components are housed in a structure made out of 80/20 aluminum extrusions.
The flowchart below details the step-by-step process for how to use the machine.
The video below shows how our machine is used to punch FTA cards and decontaminate.
For more detailed design information, please reference the Final Design tab on the left.
Performance Results
Compared to our target, our machined performed 11.9 minutes or 39.6% better.
Compared to our estimate, our machined performed 1.8 minutes or 8.5% better.
A summary of our target, estimated, and actual throughput performance is presented in the table below.