Innovation That Matters

| Photo source CATALOG

Could synthetic DNA be the future of data processing?

Computing & Tech

This platform could drastically cut the amount of energy needed by AI and machine learning

Spotted: We are living in the age of data. By 2025, global data creation and consumption are predicted to surpass 180 zettabytes – one zettabyte is equal to a staggering one trillion gigabytes. Even trying to process a tiny fraction of that data is an enormous task and conventional computing systems struggle to keep up, proving too energy-intensive and expensive. Now, CATALOG may have an answer.

The Boston-based startup has turned to the natural world for inspiration, leaning into living organisms’ ability to store and compute important information in DNA – an extremely dense and stable storage medium. CATALOG starts with pre-made synthetic DNA strands, which can be brought together in different combinations using enzymes. These molecules are encoded with data and assigned different information.

Other researchers have already seen the potential of storing information using synthetic DNA, but CATALOG is taking this one step further, developing what it claims is the first commercially viable platform for DNA-based storage and computation.

Because it uses DNA, CATALOG’s platform is capable of parallel processing, meaning it can compute and search multiple bits of data at once using relatively low amounts of energy and a small physical footprint. This would mean the technology could process data sets much more effectively, including infrequently accessed or ‘cold’ data, and identify previously overlooked patterns and trends at a scale that isn’t feasible with conventional technologies.

This could transform the way organisations and governments store and use their data, enabling vast and complicated computations to be completed without requiring overly large, expensive, and energy-intensive facilities. According to a representative of CATALOG, initial applications of the technology include AI, machine learning, data analytics, and secure computing. In particular, early use cases could be “fraud detection in financial services, image processing for defect discovery in manufacturing, and digital signal processing in the energy sector.”

Written By: Matilda Cox

Email: hello@catalogdna.com

Website: catalogdna.com

Download PDF