During their Nobel Prize Lecture on Chemistry Frances Arnold said in 2018: “Today we can read, write and edit virtually any DNA sequence, but we cannot assemble it.” That is not any longer true.
Since then, science and technology have advanced a lot that artificial intelligence has learned to assemble DNA, and with the assistance of genetically modified bacteria, scientists are on the approach to designing and producing customized proteins.
The goal is that scientists can use the design power of artificial intelligence and the technical capabilities of genome editing to engineer bacteria to act as mini-factories, producing latest proteins that may reduce greenhouse gases, digest plastics or act as species-specific pesticides.
As a Chemistry professor and computational chemist As a student of molecular science and environmental chemistry, I’m convinced that advances in artificial intelligence and genome editing make this a practical possibility.
Gene sequencing – reading the recipes of life
All living things contain genetic material – DNA and RNA – that gives the hereditary information they should reproduce and make proteins. Proteins make up 75% of human dry weight. They make up muscles, enzymes, hormones, blood, hair and cartilage. To understand proteins, you’ve to grasp numerous biology. The order of nucleotide bases in DNA, or RNA in some viruses, encodes this information, and genomic sequencing technologies discover the order of those bases.
The Human Genome Project was a global project that sequenced the complete human genome from 1990 to 2003. Thanks to rapidly improving technologies, it took seven years to sequence the primary 1% of the genome and one other seven years to sequence the remaining 99%. By 2003, scientists had the whole sequence of the three billion nucleotide base pairs that code for 20,000 to 25,000 genes within the human genome.
However, understanding the function of most proteins and correcting their malfunctions remained a challenge.
AI learns proteins
The shape of every protein is crucial to its function and is set by the sequence of its amino acids, which in turn is set by the nucleotide sequence of the gene. Misfolded proteins have the improper shape and may cause disease comparable to neurodegenerative diseases, cystic fibrosis and kind 2 diabetes. To understand these diseases and develop treatments, knowledge of protein shapes is obligatory.
Before 2016, the one approach to determine the form of a protein was to X-ray crystallographya laboratory technique that uses the diffraction of X-rays from single crystals to find out the precise arrangement of atoms and molecules in three dimensions inside a molecule. At that point, the structure of about 200,000 proteins had been determined by crystallography, at a price of billions of dollars.
AlphaFold, a machine learning program, used these crystal structures as a training set to find out the form of proteins from their nucleotide sequences. And in lower than a 12 months, this system calculated the protein structures of all 214 million genes which were sequenced and published. The protein structures determined by AlphaFold were all compiled right into a freely available database.
To effectively treat non-infectious diseases and develop latest drugs, scientists need a greater understanding of how proteins, particularly enzymes, bind small molecules. Enzymes are protein catalysts that enable and regulate biochemical reactions.
AlphaFold3published on May 8, 2024, can predict protein shapes and the sites where small molecules can bind to those proteins. In Rational drug designDrugs are designed to bind proteins involved in a signaling pathway related to the disease being treated. The small molecule drugs bind to the protein binding site and modulate its activity, thereby affecting the disease progression. By with the ability to predict protein binding sites, AlphaFold3 will improve researchers' drug development capabilities.
AI + CRISPR = composition of latest proteins
Around 2015, CRISPR technology revolutionized gene editing. CRISPR may be used to search out a selected a part of a gene, change or delete it, make the cell express kind of of its gene product, and even add a totally foreign gene as a substitute.
In 2020, Jennifer Doudna and Emmanuelle Charpentier were awarded the Nobel Prize in Chemistry “for the development of a method (CRISPR) for genome editing.” With CRISPR, genome editing, which used to take years and was species-specific, expensive and laborious, can now be done in a matter of days and at a fraction of the price.
AI and genetic engineering are making rapid progress. What was once complicated and expensive is now routine. The dream for the long run is customized proteins designed and produced through a mix of machine learning and CRISPR-modified bacteria. AI would develop the proteins, and bacteria modified with CRISPR would produce the proteins. Enzymes produced this manner could potentially breathe in carbon dioxide and methane while respiration out organic feedstocks, or break down plastics into concrete substitutes.
I consider that these ambitions usually are not unrealistic, considering that genetically modified organisms already account for two% of the US economy in agriculture and pharmaceuticals.
Two groups have created functioning enzymes from scratch, designed by different AI systems. David Baker'S Institute for Protein Design on the University of Washington developed a brand new deep learning-based protein design strategy called “family-wide hallucination”, which she used to produce a singular light-emitting enzyme. The biotech startup has now They will flowused an AI that was trained with the sum of all CRISPR-Cas knowledge to design latest functional genome editors.
If AI can learn to create latest CRISPR systems in addition to bioluminescent enzymes that work and have never been seen on Earth, there may be hope that the mixture of CRISPR and AI may be used to create other latest customized enzymes. Although the CRISPR-AI combination continues to be in its infancy, once it’s mature, it must be very useful and even help the world fight climate change.
However, one must not forget that the more powerful a technology is, the greater the risks it poses. In addition, people have not very successful in constructing nature on account of the complexity and interconnectedness of natural systems, which regularly results in unintended consequences.
image credit : theconversation.com