Keynote Address: DNA Data Storage and Near-Molecule Processing for the Yottabyte Era

on YouTube ↗01h 01m

" summary

An end-to-end system encodes bits into ~150–300 base synthetic DNA, synthesizes pools, encapsulates them, and reads with Illumina or nanopore sequencers; the team stored >1 GB and fully recovered it and estimates DNA cold storage cuts energy/carbon and water per TB-year by ~2.5–3×. Random access uses PCR-style selective copying and magnetic-bead extraction; a lab-built automated rig and digital-microfluidics stack (purple drop) moves droplets; a demo 1.6M images searched via DNA hybridization yields similarity-ranked results. Latency targets are minutes–hours.

AI-written summary · about

" chapters (22)

1. 0:00Intro
2. 2:44When Moore met Feynman
3. 4:14DNA data storage basics
4. 8:01Comparison with other media
5. 9:32Ability to make copies
6. 14:24DNA encoding
7. 15:00DNA synthesis
8. 19:16Reading DNA with sequencing by synthesis
9. 21:41Different error profile across platforms
10. 22:10DNA decoding and error correction
11. 24:32Exponential improvements in DNA data storage
12. 25:34Performance of reading and writing DNA
13. 29:22DNA storage end-to-end system w/ integrated computing
14. 34:00Searching with DNA
15. 34:39Content based media search
16. 37:24"Semantic" Hashing
17. 38:02Learning-based encoding
18. 39:31Yottabyte-scale near-molecule computing?
19. 44:07Digital microfluidics Versatile platform to implement wet lab preparation protocols
20. 45:47Affordable full-stack SW/HW digital microfluidics platform
21. 47:18Hardware, software, wetware
22. 49:51Future hybrid systems