COVID 19 is increasingly affecting elder patients. Researchers have identified drugs which when repurposed are effective to work against the coronavirus especially in elder patients. The research team led by Uhler, a computational biologist in MIT's Department of Electrical Engineering and Computer Science and the Institute for Data, Systems, and Society, and an associate member of the Broad Institute of MIT and Harvard led to the discovery.
Three steps were used by the team. First, they generated a list of drugs using the machine learning approach called autoencoder. Secondly, they map the proteins and genes that are involved in aging and SARS-COV 2 infection. Thirdly, they make use of statistical algorithms to know the causality in the gene network, thus allowing to know the upstream genes that caused cascading effects in the network, The drugs that target these upward streams of proteins and genes are the candidates for clinical trials.
The autoencoder made use of two data sets of the gene expression patterns. One of the datasets shows how expression in different cell types responded to the drugs available in the market. While the other dataset showed how expression responded towards the COVID infection. Then the list was narrowed by using the second step mentioned above. Also, the overlap areas of the maps were identified and this helped to get the correct gene network that a drug should target to attack the COVID 19 in elder patients. The correct drug would point the upstream genes of the network thus minimizing the impact created by the infection. Through the third step, it was identified that the RIPK1 is a target protein/ gene for COVID 19 drugs. A list of drugs was found that act on RIPK1. Drugs like quinapril and ribavirin are in clinical trials.