| Group |
2020-06 |
Status |
completed |
| Title |
An automated pipeline for cell growth tracking and droplet release using a microfluidic device |
| Supervisor |
Steve Shih |
| Description |
Droplet-in-channel microfluidics, is a technology, by which small picoliter droplets can be generated and manipulated, often with biological content. This has already revolutionized scientific research in a range of fields including synthetic biology, healthcare, and chemistry. A good example is the use of this technology for the encapsulation of single cells, which allows for single cell sequencing. In our lab, we have developed multiple of these devices, incorporating electrodes to facilitate droplet movement.
There are two frequent problems with single-cell based droplet microfluidic devices.
First, to sort out cell containing droplets of choice, commonly these devices rely on detecting fluorescence. Label-free detection methods are highly desired in the field.
Secondly, after sorting out a selected pool of droplets, commonly these droplets are manually collected from the device and pooled. Automated single droplet recovery into a standard recipient like a well plate is highly desired in the field.
We are currently in the process of developing a device that can sort out genetically altered filamentous fungal strains that grow better in certain conditions, for an industrial partner. For this, we wish to overcome both of these problems. We have established the device, the biology and a basic workflow, and are looking for help with automating this setup.
This project would involve designing an automated pipeline for droplet content image recognition, and work towards automated single droplet recovery. The first aim is to scan droplets over a 4hr period, perform (real-time) image recognition, detect candidates, and start a fluidic protocol as an outcome. This involves writing an image recognition pipeline for filamentous fungi in droplets and programming a scanning stage for image collection, over a 4hr period. The software needs to be able to communicate with a Python program for operation of fluidic pumps, light switches, peltiers, and, optocouplers to actuate high-voltage electrodes. The second aim is to be able to recognize single droplets that are about to exit the device, working towards a system that can automate deposition of a single droplet in a single well of a 96 well plate or other substrate. This involves prototyping a mechanical chip holder with optics.
|
| Student Requirement |
This would involve:
- Programming of XYZ microscope stage for automated image collection timelapse (Labview or MicroManager)
- Writing an image recognition program (OpenCV) that can track growth of filamentous fungi in droplets. (recognizing filamentous fungi in droplets, measuring their surface area, measuring droplet fluorescence and determining tresholds for sorting)
- Use the automated stage and droplet recognition platform, to automate the release of droplets from droplet traps
- Design of a computer driven platform that integrates automated stage scanning, image analysis and integrates flawlessly with our existing pump, electrode, Peltier and light automation (ImageJ, Micromanager, Python, OpenCV)
|
| Tools |
Prototyping (3D printing, …), Microcontrollers, Programming (matlab, python, Arduino, graphical user interfaces (Tkinter, OpenGL), OpenCV), interest in biology/microscopy/laboratory automation
|
| Number of Students |
3 |
| Students |
Daniel Gelfand
Ruslan Sinyavsky
Mariya Perlitch |
| Comments: |
Email: steve.shih@concordia.ca |
| Links: |
|