Berkeley Lab’s machine learning algorithm accelerates metabolic engineering in synthetic biology. (Image Adobestock)
Synthetic biology, like artificial intelligence (AI) machine learning, is a relatively modern field that applies emerging technologies to achieve innovation. Now scientists at Lawrence Berkeley National Laboratory (Berkeley Lab) in California have merged the two fields by creating a machine learning algorithm for synthetic biology called ART (Automated Recommendation Tool), and published their study a few weeks ago in Nature Communications.
Synthetic biology includes the design and formation of novel biological systems or components, and the redesign and production of natural biological systems. Many industries benefit from advances in synthetic biology. Examples include cosmetics, pharmaceutical drugs, vaccines, food and beverage, consumer products, agriculture, delivery plasmids, BioBrick parts, synthetic cells, bioinformatics, DNA synthesis, gene editing, oligonucleotides, chemicals, synthetic genes, and health care.
Applied synthetic biology is used to create plant-based meat substitutes. Silicon Valley-based Impossible Foods with USD 1.4 billion in funding from a myriad of venture capitalists and private investors (including Google Ventures, Serena Williams, Bill Gates, Khosla Ventures, Katy Perry, and others per Crunchbase), is a startup that uses synthetic biology to create its plant-based burgers, pork, and sausages. Specifically, scientists at Impossible Foods extracted the DNA from soy plants and placed it into genetically engineered yeast to brew up leghemoglobin, the soy protein with similarities to heme. Heme is the key molecule found in the protein myoglobin that gives meat its unique meaty flavor essence.
Artificial intelligence (AI) is flourishing anew mostly due to deep learning which is a subset of machine learning with architecture that is somewhat inspired by the biological brain. AI is gaining momentum across a wide range of purposes such as electric vehicle batteries, extreme weather predictions, flavor and fragrances, anti-vaping and bullying detectors for schools, robotics, predicting future events, and even in e-sports.