How Machine Learning and AI helped scientists reduce plastic breakdown time by 9000%.
A thread.👇 #AI#MachineLearning#Plastic
Plastics.Plastics.Plastics.
Plastic has become a necessity in today's world due to its low cost, lightweight, and long-lasting qualities.
However, it has also become a serious threat to our environment and various ecosystems.
You might be wondering now..
If plastics are causing serious harm to our environment, why can't they be recycled or, better yet, banned?
Because...
each plastic contains a different resin (a different chemical) and each resin reacts differently when reprocessed into a new item because different resins melt at different temperatures, raising the cost of recycling.
Besides that..
We do not have replacements for plastics that are as cheap, durable, and lightweight as plastic.
To help prevent increasing plastic pollution, a team of scientists at the University of Texas at Austin is working on
developing an enzyme that, with the help of a machine-learning algorithm, can reduce the time it takes for plastic to decompose from centuries to weeks.
The researchers trained a machine-learning algorithm to predict the positions of amino acids in protein structures and tested various sequences to find one that worked at different temperatures and pH levels.
And finally, they found an enzyme called
FAST-PETase, which stands for “functional, active, stable, and tolerant PETase.
It is an enzyme that can break down 51 different varieties of plastic in non-specific temperature and pH conditions in less than eight days.
"It works by breaking down plastics into their initial monomers and then once you have your original monomer, it’s as if you’re making fresh plastic from scratch, with the benefit that you don’t need to use additional petroleum resources.”(S:Hal Alper, an author on the paper)
Every day, 8 million pieces of plastic enter the oceans. Scaling up the use of FAST-PETase can assist us in conversing with our environment to make it a better place.
• • •
Missing some Tweet in this thread? You can try to
force a refresh