Longevity InTime: Autonomous AI Institute. Anti-Aging Digital Health Immortality Transhumanist AI Channel

Biomni and TusoAI Breakthrough Biomni and TusoAI have developed an open predictor of cell response to gene knockout. On July 30, Phylo presented the Biomni × TusoAI collaboration for automatic development of biological models. According to the company, the system tested around 500 variants and developed TusoPerturb, an open method for predicting changes in gene activity after CRISPR knockout of another gene. The CRISPR method allows for the knockout of a selected gene, which in turn changes the activity of many other genes. Researchers receive a long list of possible knockouts and must decide which experiment to conduct next. The genetic perturbation predictor learns from previous experiments and provides a table of expected changes in gene activity in advance. The Biomni × TusoAI system assembled the predictor from several components, testing which gene information to input, how to prepare the data, and which algorithm to use to obtain the prediction. TusoPerturb uses biological databases, including gene participation in cellular processes, connections to other genes, and cell line information. For some tasks, the method applies regression, while for others, it adds a search for similar genes and transfers their measured response. As reported in Phylo, the results of TusoPerturb are higher than previous methods, including large deep learning models, on three independent test sets. The TusoPerturb repository publishes the code, settings, and description of the path from input data to prediction. Other groups can take the same configuration, run it on supported datasets, and compare the results with their own methods. 🔗 Read original →