Evolution has evolved.

At its core, Intelligent Evolution works the same way as natural selection: start with a gene, create subtle variations across multiple generations, then select for ones best suited for a specific task or niche. The only problem with nature’s time-honored technique is that the random mutation lottery requires years to do something we can implement in a day.

And that’s where the “intelligent” part of Intelligent Evolution comes in.

Building upon Nobel Prize-winning techniques, our platform is the first to unite biology and chemistry with single-molecule biophysics and artificial intelligence, giving us a critical head start over the natural selection process (and over other directed evolution tech, too).

Our unique approach lets us identify smarter starting points and better selection criteria, which lets us uncover underlying sequence-to-function relationships more quickly. Then we use that data to refine the proprietary AI algorithms that help us generate even more specialized proteins. All told, Intelligent Evolution allows us produce enzymes that are thousands—or even millions—times more efficient than their initial form.

When it comes to evolution, the universe has nothing on us.

HOW IT WORKS

RAPID ITERATION

Large-scale protein experimentation with ultra-high-throughput assays generate massive datasets for faster improvement.

DEEP LEARNING

Scalable algorithms built on fundamental sequence-to-function principles identify critical relationships within datasets.

NOVEL DESIGN

Industry-leading application of single-molecule biophysics enhance protein design and characterization.

EVIDENCE IN LITERATURE

PREDICTING STRUCTURE

“...the first, to our knowledge, computational approach capable of predicting protein structures to near experimental accuracy in a majority of cases.” (Jumper et al., 2021)

Read the article >

RECYCLING WITH ENZYMES

“...the biodegradation of plastics by specialized bacteria could be a viable bioremediation strategy.” (Yoshida et al., 2016)

Read the article >

DESIGNING PROTEINS

“...deep networks trained to predict native protein structures from their sequences can be inverted to design new proteins.” (Anishchenko et al., 2021)

Read the article >

USING MICROFLUIDICS

“...femolitre droplets can be used as microreactors for molecular biology with volumes one billion times smaller than conventional microtitre plate wells.” (Leman et al., 2015)

Read the article >