Why store data on DNA?

Size: The entire internet one room

Internet big + AI even bigger. Need smaller solutions, running out of space. Lorem ipsum dolor sit amet consectetur adipiscing elit. Quisque faucibus ex sapien vitae pellentesque sem placerat. https://www.science.org/content/article/dna-could-store-all-worlds-data-one-room

Age: A true long-term storage solution

DNA can last for thousands of years (cite and compare this) Lorem ipsum dolor sit amet consectetur adipiscing elit. Quisque faucibus ex sapien vitae pellentesque sem placerat.

Energy costs

Much lower than traditional methods. Lorem ipsum dolorem sit amet consectetur elit. Lorem ipsum dolor sit amet consectetur adipiscing elit. Quisque faucibus ex sapien vitae pellentesque sem placerat.

Relevancy: Relevant as long as life as we know it is

Data loss from outdated mediums. Who will know how to use a VHS in 100 years? Lorem ipsum dolor sit amet consectetur adipiscing elit. Quisque faucibus ex sapien vitae pellentesque sem placerat. In id cursus mi pretium tellus duis convallis. Tempus leo eu aenean sed diam urna tempor.

https://wyss.harvard.edu/news/save-it-in-dna/

Background

Bases: How DNA stores information

The main part of DNA that we'll be focusing on are its nitrogenous bases, the way DNA stores information. These four bases are 'Adenine', 'Guanine', 'Cytosine' and 'Thymine', also often referred to as 'A', 'G', 'C', and 'T'. These bases are complimentary: Adenine binds with Thymine, and Guanine binds with Cytosine.

Codons: How DNA is read

DNA is read as triplets of bases called 'codons', that each encode either a command (like 'start reading' or 'stop reading'), or an amino acid, the building blocks of proteins.

Notice the 'U' base in the above chart. This is because the codon chart is using mRNA, not DNA. mRNA is a DNA-like substance that has the base Uracil (U) instead of Thymine, as well as a slightly differnet sugar base. When DNA is copied to create amino acids, mRNA, or 'Messanger RNA' stores the copied information and transports it to the ribosome for it to be read and translated into a string of amino acids (aka a polypeptide chain).

PCR: How we read DNA sequences

PCR, or 'Polymerase Chain Reaction', is a technique to increase the amount of ('amplify') a specific sequence of DNA, especially to make it easier to read.

Short strands of DNA that are complimentary to the desired sequence, called 'primers', are used to identify the correct sequence of DNA. These primers bind to the end of the sequence, and then DNA polymerase adds nucleotides to the ends of the primers, duplicating the DNA sequence.

https://www.britannica.com/science/polymerase-chain-reaction

Encoding Schemes

'Naive' Approach: Base-4 Encoding

Lorem ipsum dolor sit amet consectetur adipiscing elit. Quisque faucibus ex sapien vitae pellentesque sem placerat. In id cursus mi pretium tellus duis convallis. Tempus leo eu aenean sed diam urna tempor.

Codon-Based Encoding

Quisque faucibus ex sapien vitae pellentesque sem placerat. In id cursus mi pretium tellus duis convallis.

Huffman Code & Rotating Base-3 Encoding

Quisque faucibus ex sapien vitae pellentesque sem placerat. In id cursus mi pretium tellus duis convallis.

Washington University & Microsoft: https://homes.cs.washington.edu/~luisceze/publications/dnastorage-asplos16.pdf

Error Mitigation

Common Error Mitigation Techniques

Constrained Coding VS Unconstrained Coding

The following types of sequences cause issues with the synthesis & sequencing processes:

• Homopolymers: repeated bases, like 'BBBB'
• Unbalanced GC content: roughly half of the bases should be either guanine or cytosine.
• 'Hairpin' issues: sequences that contain complimentary parts (like "AGC TCG") will bond to themselevs, causing issues with PCR
• Matching suffixes & prefixes

It appears that there are two main approaches to avoiding these sequences. There is 'constrained coding', which specifically generates sequences that avoid these issues, and then 'unconstrained coding', which simply creates pseudo-random sequences.

Weindel et. al: https://ieeexplore.ieee.org/document/11164904

Further Considerations

Future-Proofing: Can we make the encoding accessible to future civilizations?

Lorem ipsum dolor sit amet consectetur adipiscing elit. Quisque faucibus ex sapien vitae pellentesque sem placerat. In id cursus mi pretium tellus duis convallis.