Scientists have developed a synthetic TnpB enzyme using artificial intelligence that outperforms natural reference enzymes in gene editing [1, 2].
This advancement is significant because it allows researchers to create tools with properties not found in nature. By moving beyond naturally occurring CRISPR proteins, scientists can potentially increase the efficiency and specificity of genetic modifications [1, 5].
The new synthetic enzymes, called SynTnpBs, were detailed in a study published this month in the journal Science [1, 2]. The research focused on expanding the existing CRISPR toolbox by designing nucleases from the ground up rather than modifying existing biological sequences [1, 3].
Natural CRISPR systems rely on proteins that evolved for specific bacterial defense mechanisms. However, these natural constraints can limit how effectively a tool works in a laboratory or clinical setting. The AI-driven approach allows for the creation of enzymes that are optimized for human-led editing tasks, an evolution that would take millions of years in the wild [1, 5].
By utilizing AI to predict protein folding and function, the researchers were able to engineer SynTnpBs to be more effective than the natural TnpB enzymes they were modeled after [1, 2]. This process demonstrates a shift toward fully synthetic biology, where the software defines the tool before the laboratory creates the molecule [3].
While the study focuses on the technical success of the SynTnpBs enzymes, the broader implication is the ability to customize gene-editing tools for specific genomic targets [1, 5]. This flexibility could lead to more precise treatments for genetic disorders by reducing off-target effects, and increasing the success rate of the edit [1].
“AI-designed synthetic TnpB enzymes outperform natural reference enzymes.”
The transition from discovering natural enzymes to designing synthetic ones marks a pivotal shift in biotechnology. By using AI to bypass the limitations of natural evolution, researchers can create highly specialized tools tailored for specific medical or agricultural applications. This increases the likelihood of developing safer, more efficient gene therapies that are less prone to the errors associated with natural CRISPR proteins.
