Document Document

Project Description

Background



With the rapid development of society, the amount of data generated is increasing exponentially. Data explosion is coming, and we are reaching the point where traditional media, such as hard drive disks, become difficult to cope with the increasing demands for data storage.

As a new medium, DNA has the advantages of high density, long-term durability, parallel access and low energy consumption.

Many scientists have made great efforts to develop DNA storage technology, including encoding movie clips in to E.coli [1]. Recently, Yingjin Yuan's team stored the Chinese Dunhuang murals in DNA, and made them can be preserved for thousands of years at room temperature and for more than 20,000 years under 9.4°C[2].

Inspired



In 2017, George Church's team of the Wyss Institute divided 36×26 pixels of "Horse in Motion" into five frame and stored them in Escherichia coli [1]. In fact, each frame in this video doesn't change much, but each frame is stored with re-synthesized DNA. This inspired us, like keyframe animation, whether we can generate and store a video by editing only one frame. Therefore, we took Micro Venus, the first image stored in DNA [3], as the first keyframe, stored and modified it into a video clip.

In addition, we wanted to make further attempts to store and edit information besides pictures, such as music, so that the music without rhythm and tones could be turned into a piece of beautiful music. We were going to turn a nonsense song into an Ode to Joy.



Inspired by multi-site editing technology via ultra-long gRNA arrays developed by Prof. Wu's team, our team proposed the concept of DNA re-editing. By using the characteristics of the base editors that they are able to achieve conversion of C to T, we could achieve large-scale modification of encoded information simultaneously.


What have we done




Our team designed two information encoding systems to test our ideas.

One is to transform pixel images into DNA sequences. We stored a black and white pixel image, with “0” for white and “1” for black. This allowed us to change the color of a pixel by editing only one site, allowing us to modify the entire image in a limited amount of DNA. In order to modify the image with higher efficiency, we developed the second-generation image encoding program, which can obtain seven different images after modifying a sequence simultaneously.

The other one is to encode music. By modify its corresponding DNA information sequence, we could change the pitch of the music. By designing different variations of the four coding units, we were able to make arbitrary changes in pitch and length of one tone.

Details of this section can be viewed in model click here


We used these two encoding systems to artificially design two information plasmids. We designed specific targeting sites for data sequence. Then,the data plasmid and gRNA array plasmid were both transformed into cells that were already been inserted with the CBE (cytosine base editor). Gene editing occurred when CBE were induced. After that, PCR products were sequenced and we aligned them with original data sequence. Finally, we used our decoding program to transform the sequencing results back to images or music.

Details of this section can be viewed in Design and Results
click here to Designclick here to Results

Innovation



The innovations of our project:
1. We designed two sets of coding program for image and music, which convert image and music information into DNA sequences.
2. Through the Micro Nuwa multi-site editing system, we have been able to edit DNA sequences with encoded information for up to 27 sites in a single colony.
3. The edited sequence was input into our decoding and correction program, which can be transformed into another images and another piece of music as our expectation.
Micro nuwa serves as a powerful tool efficiently modifying the information stored in DNA.

References


[1] Shipman S L, Nivala J, Macklis J D, et al. CRISPR-Cas encoding of a digital movie into the genomes of a population of living bacteria[J]. Nature, 2017, 547(7663): 345-349.
[2] Song L, Geng F, Gong Z Y, et al. Robust data storage in DNA by de Bruijn graph-based de novo strand assembly[J]. Nature communications, 2022, 13(1): 1-9.
[3] Davis, Joe. Contemporary Art and the Genetic Code. Art Journal, vol. 55, no. 1,1996, pp.
70-74.