DNAPrintingPipeline

This is a simple software designed so that you can print an image on gel with DNA as your ink:)

This computational pipeline is implemented into a user-friendly platform that gives you a PCR protocol to generate the DNA fragments you need for each lane in gel electrophoresis to produce the image on gel.

Repository Developers: Tom Fu, Kariessa Schultz.

Wet Lab Members: Roya Amini-Naeini, Michelle Lee.

Running the pipeline

step 1. make sure you have the right libraries downloaded

This project uses the following python library:
Panda, PIL, numpy, math, os(operating system control)

To make sure you have the right libraries downloaded, in your terminal, cd to enter the current directory (DNAPrintingPipeline), type pip3 install -r requirements.txt. If your terminal says that you don't have pip3 installed, try using pip instead. Note that math and os should be built in python dictionaries without the need of installing.

Install kivy with python -m pip install kivy. Adjust python into python3 according to your os.

step 2. run the GUI

To run this GUI, clone this repository and enter the directory.

In the terminal, enter python DNAPrintingStation.py. A Kivy viewing panel should pop up. Note that our pipeline works for Mac and Linux with tested stable behaviours, but Kivy has failed before on Windows in our testing, as seen below. You might want to use Macbooks or a Linux computer if your Windows computer does not work. As before, adjust python into python3 depending on your os. You should run our software using the same Python version that you used to install Kivy.
In the pop up viewing panel, you should see the first screen that tells you that this is indeed our DNA printing Project software. Click on Select an image to enter the software.
You will now see an image selector tab. Select the image you want to print on gel from your system. We provide a set of pictures in the ./testImages directory. Note that we will recommend simpler pictures in terms of colour - the image would be printed the best if there are only two colours.
Then you should see an image confirmation page asking you to confirm the image (currently skipped).
You will now see a preview of what your picture would look like on gel, with [ ]s representing blank space not printed with DNAs and [X]s representing the space printed with DNAs, such as the one below.
If you would like, click on manual adjustment to manually adjust the output gel image if you are not satisfied with the automatically generated image. Note that you should do this by deleting the blank spaces inside the brackets with Xs, not directly typing in Xs or deleting blank spaces. Then you should be able to preview again by clicking on the Save adjustment-preview again button.
If you are happy with the preview, click on Happy with it? Print! to get your protocol. Your output PCR protocol should be in the ./output directory as a text file called protocol.txt which explains the primer pairs needed for each lane.
If you are still not happy with the preview, feel free to go back to manual adjustment again to change anything!

A note about settings in the GUI

Note that we have a setting tab on the first screen when you enter the GUI. You can enter the tab by clicking on the settings button in the front page of the software.

That is where you can change the lane numbers according to the number of lanes on your actual gel or the gel box you are using. Note that since our system uses the primer pairs we designed - we have 39 pairs of primers for now that form linearly spaced DNA sequences, the vertical dimension can only be 39 for now. So when you choose the size of the gel box (i.e. the lane number), please make sure to choose an appropriate number so that the image does not look overly squished. Lane number is set to 20 by default.

Most recent update: Minimum viable product finished (August 18, 2020)

Implemented by Kariessa and Tom as well as experimental/conceptual support from Roya, the functionalities the GUI has include:

Take in an input image
Allow for manual adjustments on lane numbers
Allow for user confirmation on image selection
Allow for gel simulation to preview the printed gel
Allow for manual adjustment on what the printed gel should look like
Form protocol for PCR that allows DNA to be printed on gels in the shapes we want

Future optimisation

Roya is conducting wet-lab experiments to corroborate the effectiveness of our software.

In addition, thoughts on future optimisation include:

Print with more than 2 colors
Increase the image dimension from 2D to 3D
Use the OpenTron robot to automise the pipeline.

Acknowledgement

This project is a collaborative effort of Tom Fu, Kariessa Schultz, Roya Amini-Naini, and Michelle Lee at the Harvey Mudd College BioMakerspace. The Harvey Mudd faculty advisor for this project is Dr. Dan Stoebel.

This project is part of the intellectual property of the Harvey Mudd College BioMakerspace. Please notify the space at tfu@g.hmc.edu for consent before using any of the materials in this repository. Please cite us if you use our repository for academic sharing purposes.

The funding support of the Harvey Mudd College BioMakerspace comes from Harvey Mudd College Office of Community Engagement and the college's Shanahan Fund.

The image analysis part of thie pipeline is done with the generous help of Bowen. Some source files are from Roya and Max.

Cheers,
Tom Fu on behalf of the Harvey Mudd College Biomakerspace
August 18, 2020

tommyfuu/DNAPrintingPipeline