The University of Minnesota recently reported that metformin, a diabetes drug I happen to take, which is also a candidate to be classed as a senolytic, reduces the chances of developing long Covid, and also that ivermectin and fluvoxamine do not. But the interesting bit for me?
Metformin’s ability to stop the virus was predicted by a simulator developed by U of M Medical School and College of Science and Engineering Biomedical Engineering faculty. The model has been highly accurate to date, successfully predicting, among others, the failure of hydroxychloroquine and the success of remdesivir before the results of clinical trials testing these therapies were announced. [University of Minnesota/Medical School]
OK, so tell me about the simulator.
The biophysics-based model simulated [COVID-19] on a molecular and cellular level so the trial team could screen potential treatments computationally long before they were given to participants. [University of Minnesota/Biomedical Engineering]
OK, cool. That’ll take a bit of horsepower, I’m sure. But this surprised me:
Meanwhile, other members of the clinical trial team noticed metformin’s potential through their own methods. In April 2020, natural language processing of SARS-CoV-1 and SARS-CoV-2 performed by Christopher Tignanelli, MD, and Rui Zhang, PhD, Director of the University’s Natural Language Processing/Information Extraction Program identified a class of drugs known as mTOR inhibitors — which includes metformin — that had strong potential to stop the viral life cycle.
So what’s going on here? Following the link:
The Natural Language Processing/Information Extraction (NLP/IE) Program is a teamof investigators, postdoctoral researchers, developers, and students working together since 2009 advancing capabilities to process, extract, and encode information from unstructured biomedical and clinical texts, including clinical notes from the electronic health record and biomedical literature. Current active areas of NLP/IE research for our group include redundancy detection in clinical texts; biomedical semantic similarity and relatedness measures; acronym, abbreviation, and symbol disambiguation; semantic role labeling; automated monitoring of adverse drug events; literature-based discovery for drug repurposing; algorithms to extract phenotyping; tools for interoperability and integration of NLP systems; and specialized modules for different types of clinical texts.
In other words, the body of knowledge making up biomedical engineering is so large and dis-coordinated that an expert, a professor, no longer carries the whole mess around in their head. This tool reformulates the mess, digging out useful information and tentative conclusions for use by researchers.
Goodness. And cool.
