Scientists at Oak Ridge National Laboratory have developed CRISPR Cas9 technology for genome editing in microorganisms using quantum biology and explainable artificial intelligence. This breakthrough allows for more precise genetic modification of microorganisms, opening up possibilities for the production of renewable fuels and chemicals.
Oak Ridge National Laboratory’s quantum biology and AI research has significantly improved the efficiency of CRISPR Cas9 genome editing in microorganisms and contributed to renewable energy development.
Scientists at Oak Ridge National Laboratory (ORNL) are leveraging expertise in quantum biology, artificial intelligence, and bioengineering to improve the We’ve improved the way the CRISPR Cas9 genome editing tool works.
CRISPR is a powerful tool for bioengineering, used to modify the genetic code to improve the performance of organisms or correct mutations. CRISPR Cas9 tools rely on a single, unique guide. RNA This instructs the Cas9 enzyme to bind and cut the corresponding target site in the genome. Existing models to computationally predict effective guide RNAs for CRISPR tools were built on data from only a few models. seedthe efficiency is weak and inconsistent when applied to microorganisms.
CRISPR research focused on microorganisms
“Many CRISPR tools have been developed for mammalian cells, Drosophila melanogaster, or other model species. Few have been targeted for microorganisms, where chromosome structure and size vary widely,” says ORNL’s Synthetic Biology Group. said Carrie Eckert, leader of “We had observed that models for designing the CRISPR Cas9 machinery behaved differently when working with microorganisms, but this study validates what we knew anecdotally. Masu.”
![Quantum biology offers better gene editing tools](https://scitechdaily.com/images/Quantum-Biology-Informs-Better-Gene-Editing-Tool-777x494.jpg 777w,https://scitechdaily.com/images/Quantum-Biology-Informs-Better-Gene-Editing-Tool-400x254.jpg 400w,https://scitechdaily.com/images/Quantum-Biology-Informs-Better-Gene-Editing-Tool-768x488.jpg 768w,https://scitechdaily.com/images/Quantum-Biology-Informs-Better-Gene-Editing-Tool-1536x977.jpg 1536w,https://scitechdaily.com/images/Quantum-Biology-Informs-Better-Gene-Editing-Tool-2048x1302.jpg 2048w)
![Quantum biology offers better gene editing tools](https://scitechdaily.com/images/Quantum-Biology-Informs-Better-Gene-Editing-Tool-777x494.jpg 777w,https://scitechdaily.com/images/Quantum-Biology-Informs-Better-Gene-Editing-Tool-400x254.jpg 400w,https://scitechdaily.com/images/Quantum-Biology-Informs-Better-Gene-Editing-Tool-768x488.jpg 768w,https://scitechdaily.com/images/Quantum-Biology-Informs-Better-Gene-Editing-Tool-1536x977.jpg 1536w,https://scitechdaily.com/images/Quantum-Biology-Informs-Better-Gene-Editing-Tool-2048x1302.jpg 2048w)
ORNL scientists have developed a method to improve the accuracy of the CRISPR Cas9 gene editing tool used to modify microorganisms for the production of renewable fuels and chemicals. The research leverages the lab’s expertise in quantum biology, artificial intelligence, and synthetic biology.Credit: Philip Gray/ORNL, U.S. Department of Energy
To improve the modeling and design of guide RNAs, ORNL scientists sought to better understand what is happening at the most fundamental level in the cell nucleus, where genetic material is stored. They turned to quantum biology, a field that bridges molecular biology and quantum chemistry, to study how electronic structure affects the chemical properties and interactions of nucleotides (molecules that form the building blocks of living things). did. DNA and RNA.
Erica Plates, a computational systems biologist at ORNL, said that the way electrons are distributed within a molecule may affect reactivity and conformational stability, including the possibility that the Cas9 enzyme-guided RNA complex will bind effectively to microbial DNA. He said it affects sexuality.
Utilizing Explainable AI in CRISPR research
Scientists have built an explainable artificial intelligence model called iterated random forest. They trained their model on a dataset of about 50,000 guide RNAs targeting the genome. Escherichia coli The approach described in the journal analyzes bacteria while also taking into account their quantum chemical properties. Nucleic acid research.
This model revealed important features regarding the nucleotides that allow for better selection of guide RNAs. “This model helped us identify clues about the molecular mechanisms underlying guide RNA efficiency. This provided a rich library of molecular information to help improve CRISPR technology,” said Prates. says.
ORNL researchers validated their explainable AI model by conducting CRISPR Cas9 cleavage experiments. Escherichia coli Features a large group of guides selected by model.
Jaclyn Noshey, a former ORNL computational systems biologist and lead author of the paper, said scientists can use explainable AI rather than deep learning models rooted in uninterpretable “black box” algorithms. , said they were now able to understand the biological mechanisms that lead to their results. .
“Given our knowledge of the incompatibility of models trained in different fields, we wanted to deepen our understanding of guiding design rules to achieve optimal cutting efficiency with a focus on microbial species. . [biological] It’s a kingdom,” Noshei said.
An explainable AI model with thousands of features and repeatability was trained using the Summit supercomputer at ORNL’s Oak Ridge Leadership Computing Facility (OLCF), a DOE Office of Science user facility.
Eckert said her synthetic biology team is collaborating with computational science colleagues at ORNL to take what they learned with the new microbial CRISPR Cas9 model and use lab experiments and data from a variety of microbial species to further develop it. He said he plans to improve it.
Advances in CRISPR Cas9 tools for diverse species
Considering quantum properties opens the door to improvements in Cas9 guides of all species. “This paper has implications across the human scale,” Eckert said. “If you’re looking at any kind of drug development, for example using CRISPR to target specific regions of the genome, you need the most accurate model to predict those guides. ”
Improving the CRISPR Cas9 model provides scientists with a high-throughput pipeline for linking genotype to phenotype, or genes to physical traits, in a field known as functional genomics. This research will impact efforts at the ORNL-led Center for Bioenergy Innovation (CBI), such as improving bioenergy feedstock plants and bacterial fermentation of biomass.
“This study significantly improves the prediction of guide RNAs,” Eckert said. “The more we can understand the biological processes going on and incorporate more data into our predictions, the better our goals will be, and the more accurate and faster our research will be.”
“The main goal of our research is to improve our ability to use CRISPR tools to predictively modify the DNA of more organisms. “This represents an exciting advance toward understanding how we can avoid ‘mistakes’,” said ORNL’s bioanalytical chemist who leads the DOE Genome Science Program’s Safe Ecosystem Engineering and Design Science Focus Area. Paul Abraham says. , or SEED SFA, supported CRISPR research. “We are interested to see how much these predictions can improve as we continue to generate additional training data and leverage explainable AI modeling.”
Reference: “Quantum biological insights into CRISPR-Cas9 sgRNA efficiency from explainable AI-driven feature engineering” Jaclyn M Noshay, Tyler Walker, William G Alexander, Dawn M Klingeman, Jonathon Romero, Angelica M Walker, Erica Prates, Carrie Eckert, Stephan Irle, David Kainer, Daniel A Jacobson September 20, 2023; Nucleic acid research.
DOI: 10.1093/nar/gkad736
Co-authors of this publication include ORNL’s William Alexander, Dawn Klingeman, Erica Prates, Carrie Eckert, Stephan Irle and Daniel Jacobson. Tyler Walker, Jonathan Romero, and Angelica Walker of the Bredesen Center for Interdisciplinary Research and Graduate Education at the University of Tennessee, Knoxville; Jaclyn Noshea and David Kiner were previously at ORNL and currently at Bayer and the University of Queensland, respectively.
Funding for this project was provided by SEED SFA and CBI, part of the DOE Office of Science’s Biological and Environmental Research Program, ORNL’s Laboratory-Initiated Research and Development Program, and OLCF and Compute’s High Performance Computing Resources. and Data Environment for Science, both supported by the Office of Science.