Scientists at the Lawrence Berkeley National Laboratory, or Berkeley Lab, have designed an Automated Recommendation Tool, or ART, that seeks to revolutionize the field of bioengineering by employing a systematic and automated approach to synthetic biology.
The field of synthetic biology entails generating “novel valuable molecules,” such as biofuels or anti-cancer drugs, according to a study conducted by Berkeley Lab researchers. Scientists face limitations, however, because of the years-long effort it requires to understand and manipulate a cell’s biochemistry to produce a desired outcome.
ART pairs synthetic biology with machine learning algorithms to allow for the automated metabolic engineering of a cell’s DNA and the resultant creation of cells, according to a Berkeley Lab press release.
“People are making synthetic milk, burgers, spider silk, collagen and biomaterials,” said Héctor García Martin, study co-author and staff scientist in Berkeley Lab’s Biological Systems and Engineering, or BSE, division. “This is just a tool that makes bioengineering easier because it really takes out the problem of ingesting all that data and deciding what the next step is.”
By predicting how changes in a cell’s biochemistry will affect its behavior and making subsequent “recommendations for the next engineering cycle,” ART accelerates synthetic development, allowing it to occur in a span of weeks or months, according to the press release.
Researchers from Berkeley Lab and their collaborators at the Technical University of Denmark demonstrated ART’s capabilities by using it to guide the engineering of the amino acid tryptophan, according to García Martin.
By acquiring experimental data from five genes, ART was trained to recommend a design that increased tryptophan production by 106% in yeast, according to the press release.
Tijana Radivojević, study co-author and BSE data scientist, emphasized the need for more investment and greater access to datasets in order to unlock the extent of ART’s capabilities.
“The algorithm itself is agnostic to the specific output or response of the system,” Radivojević said. “As long as we have the proper training datasets that define the inputs for the system and the associated response, we could apply this tool.”
Researchers aim to apply ART so that it may produce an environmentally beneficial impact, according to García Martin. They are in the process of creating biofuel molecules based off of biological designs generated by the algorithm.
With the necessary data to feed algorithms such as ART, there is potential for the generation of cells that can produce biofuels at “commercially viable levels,” according to García Martin. He added that this would displace all fossil fuels currently in use, which are emitting greenhouse gases and “significantly” affecting the climate.
Scientists are continuing to search for ways to make bioengineering a fully automated process, García Martin said.
Machine learning presents a route by which bioengineers may more efficiently arrive at “meaningful conclusions” with less experimentation, said Kenneth Workman, study co-author and UC Berkeley junior, in an email.
“An emerging field of synthetic biology is using deep learning to design synthetic DNA sequences from scratch to engineer life in any which way,” Workman said in the email.