Cell heavyweight! Chinese scientist David R. Liu develops a new way to predict gene editing results!

June 2020 /BiovalleyBIOON / -- Although base editors are widely used to achieve fixed-point mutations, the factors that determine the results of base editing have not been well understood and often require testing to verifyResearchers from the Broad Institute, Harvard University, the Massachusetts Institute of Technology and the Brigham and Women's Hospital and Harvard Medical School, led by Chinese scientist David RLiu, developed a machine learning model to predict the results of editing, without experiments, published in Cell, in a study entitled "The Sorres of The Base Editing Outcomes from Target Library And Learning Machine."picture source: Cell
highlights of the research paper include:the accuracy and efficiency of the underlying editing results are often incongruous, while the Machine Learning Model (BE-Hive) can accurately predict basic editing efficiency and editing patterns;basic editor engineering can increase and reduce offside editing; andaccurately corrected 3,388 pathogenic SNVs, many of which were previously considered stubbornin the study, the researchers characterized the sequence strain and activity of 38,538 genes in mammalian cells with 11 cytosine and pyridine editing (CBEs and ABEs) and then used the results to train BE-Hive, a machine learning model that accurately predicts the results of the underlying editing genotype (R 0.9) and efficiency (R 0.7)based on this, the researchers corrected 3,388 disease-related SNVs with an accuracy of 90 percent, including 675 correctly predicted unedited allels with bystander nucleotidesthe researchers found the determinants of previously unpredictable C-to-G or C-to-A editing, and used the findings to correct the coding sequences of 174 pathogenic transposing SNVs with an accuracy of up to 90%finally, the researchers used BE-Hive's insights to design new CBE variants to adjust editsin general, these findings illustrate base editing, support editing previously difficult targets, and provide improved editing capabilities for the new basic editor(BioValleyBioon.com)References: David R Liu et al.
The S of The Base Editing Outcomes from Target Library Analysis and Learning Machine.
Cell DOI: https://doi.org/10.1016/j.cell.2020.05.037