The three-dimensional analysis of cell types and their locations by spatial transcriptomics provides key information of their interactions within tissues or organs. Based on this technology, scientists at the Wellcome Sanger Institute have developed an AI tool called Nichecompass, which shows a comprehensive view of the cancer microenvironments, the different cells, their locations, and how they communicate with each other through different molecules inside the tumor. This AI could process data in an hour and compare samples before and after a treatment.

Scientists at the Institute of Cancer Research (ICR) in the U.K. are developing a technology that analyzes, in vitro, how the 3D morphology of cancer cells changes when exposed to a compound, using AI to predict their response to new treatments. The researchers estimate that their methodology could accelerate drug development by 6 years, by ruling out unsuccessful drugs and thus reducing the number of preclinical trials.

A new version of Evo, the AI developed at the Arc Institute that can be used to design genomes as long as that of a bacterium, has been retrained with the DNA sequences of three domains of life – viruses, bacteria and eukaryotes.

Tevogen Bio Holdings Inc. has expanded its relationship with Microsoft Corp. to broaden their AI-focused collaboration and build its Predictcell technology for predictive precision T-cell targeting.

Immunoprecise Antibodies Ltd. (IPA) has developed a new class of GLP-1 therapies entirely through AI, designed to enhance efficacy, safety, therapy longevity and patient satisfaction for the treatment of diabetes.

Following Nobel Prize-winning chemist David Baker’s recipe for cooking an antidote to cobra venom using artificial intelligence (AI) could be faster and more effective than currently available methods. The ingredients and steps can be found in a new study published by the University of Washington (UW) scientist in collaboration with the Technical University of Denmark. They are ready for the next steps in preclinical trials.

Artificial intelligence (AI)-based models developed by a team of international researchers were able to identify ovarian cancer in ultrasound images more accurately than humans. Results from a study published in Nature Medicine showed that the AI models achieved an accuracy rate of 86.3%, compared to 82.6% for the experts and 77.7% for the non-expert examiners.

An international consortium of thousands of scientists is creating the Human Cell Atlas, a three-dimensional map of all the cells in the body. The goal is to understand all the cells that make up human tissues, organs and systems, which will enable multiple medical applications. This collection of cell maps is openly available for navigation at single-cell resolution, identified through omics analyses that reveal the tridimensional distribution of each cell.

The development of new machine learning tools like Alphafold and Rfdiffusion has allowed scientists to predict the structure of proteins and design them for drug discovery purposes, among other uses. Now, scientists at the Arc Institute have created Evo, an AI model that generates DNA sequences and estimates their interaction with other molecules at single-nucleotide resolution, scalable to an entire genome.

Many studies have linked the presence of specific bacteria to various diseases. But a general overgrowth of gut bacteria can be a symptom of different conditions, including colorectal cancer and inflammatory bowel disease. A study counting gut microbiome proposes that microbial load, rather than the disease, could explain the presence of certain pathogens.