Monday, February 27, 2023

Science bits: AI designs new enzymes; A neuroscience advance; Hackers hacking

This one could be a game changer: Since the 1980s humans have been tinkering with changing genes and enzymes to make them work better or differently. Directed evolution is one way to generate new enzymes. Genetic engineering has advanced to the point of being able to engineer humans. Now, artificial intelligence software is starting to design new enzymes. Sci Tech Daily writes
A natural language model has jumpstarted the process of protein design by creating active enzymes.

Researchers have developed an AI system that can generate artificial enzymes from scratch. In laboratory experiments, some of these enzymes demonstrated efficacy comparable to natural enzymes, even when their artificially created amino acid sequences greatly deviated from any known natural protein.

The experiment shows that natural language processing, initially created for reading and writing language text, can grasp certain fundamental concepts of biology. The AI program, known as ProGen, was developed by Salesforce Research and employs next-token prediction to construct artificial proteins from amino acid sequences.

Scientists said the new technology could become more powerful than directed evolution, the Nobel-prize-winning protein design technology, and it will energize the 50-year-old field of protein engineering by speeding the development of new proteins that can be used for almost anything from therapeutics to degrading plastic.

“The artificial designs perform much better than designs that were inspired by the evolutionary process,” said James Fraser, Ph.D., professor of bioengineering and therapeutic sciences at the UCSF School of Pharmacy, and an author of the work, which was recently published in Nature Biotechnology. 

“The language model is learning aspects of evolution, but it’s different than the normal evolutionary process,” Fraser said. “We now have the ability to tune the generation of these properties for specific effects. For example, an enzyme that’s incredibly thermostable or likes acidic environments or won’t interact with other proteins.”  
With proteins, the design choices were almost limitless. Lysozymes are small as proteins go, with up to about 300 amino acids. But with 20 possible amino acids, there are an enormous number (20^300 - twenty to the 300th power) of possible combinations. That’s greater than taking all the humans who lived throughout time, multiplied by the number of grains of sand on Earth, multiplied by the number of atoms in the universe. (🥴!)

Given the limitless possibilities, it’s remarkable that the model can so easily generate working enzymes.
This is an area of application for AI that could revolutionize medicine. It's still too early to predict the long-term impacts. Buy my gut tells me this will turn out to be a big deal within the next couple of years. This will play out pretty fast, for better or worse.

------------------------------------------------------------------------
------------------------------------------------------------------------

An advance in understanding the brain: Medical Express writes:
For the first time, researchers record long-term electrical activity in a single brain cell

.... scientists need to understand how individual brain cells contribute to a larger network of brain activity and what role each cell plays in shaping behavior and overall health. Until now, it's been difficult to get a clear view of how brain cells in living animals behave over extended periods of time. 

But Jia Liu's group at the Harvard John A. Paulson School of Engineering and Applied Sciences (SEAS) has developed an electronic implant that collected detailed information about brain activity from a single cell of interest for more than a year.

"This research solves a fundamental issue—the challenge of creating a brain-electronic interface that does not disturb brain function or degrade over time," says Liu.

"Maybe one day it's cold and gray outside, and you feel unhappy and in a bad mood. Another day, it's sunny and you're on the beach and you're in a great mood. How those representations change in the brain cannot be studied by current technology because we haven't been able to stably track activity from the same neuron," he says. "This research completely overcomes that limitation. It's the beginning of a new era of neuroscience."
The researchers intend to transmit brain activity in real time from the brain to an artificial neural network in a computer for analysis. Researchers will look at how the mesh nanoelectronic sensors can be used to study phenomena such as "neural representation", which is how the brain generates images we perceive via vision.

This is a significant advance in brain-machine interface technology. At some point, computers and brains will be communicating. That could lead to major knowledge advances in hard to study areas such as consciousness, sentience, and the will to live. 

------------------------------------------------------------------------
------------------------------------------------------------------------

Technology is a two-edged sword: As is usual these days, morally rotted people rush in and misuse new technology that was designed to be helpful and good. Mashable writes:
Scammers are spoofing ChatGPT to spread malware

No new product can be called successful online until the scammers show up

ChatGPT has blown up in just a few months' time, becoming the fastest growing app of all time.

So, of course, hackers are already weaponizing the popularity of OpenAI's artificial intelligence chatbot in order to scam internet users.

Cybersecurity researchers have already uncovered hundreds of recently registered(Opens in a new tab) domains utilizing the term "ChatGPT." While not all of these domains will be weaponized for nefarious purposes, some of them already are being used in that way.  
.... as first reported by Bleeping Computer, one such website "chat-gpt-pc.online" attempted
to convince visitors to its page that ChatGPT was offered as a downloadable local application for Windows. Alvieri found that this download would inject users with the RedLine information-stealing malware. Essentially, this malware steals stored information in users' applications, such as their web browser. For example, if a user has Google Chrome store their passwords or credit card information, this malware can pull the data and send it to the hacker. 
Sigh, this is why we have so many bad things these days. 

No comments:

Post a Comment