Scientists used machine learning to find drugs already on the market that could also combat COVID-19 in elderly patients.
“Making new drugs takes forever,” said study co-author Caroline Uhler, a computer biologist at MIT. “Really, the only convenient option is to reuse existing drugs.”
The study team looked for potential treatments, analyzing changes in gene expression in lung cells caused by both the disease and aging.
Uhler said that this combination can help medical experts find drugs to test in older people:
We need to look at aging together with SARS-CoV-2 – what are the genes at the intersection of these two pathways?
THE researchers sought to answer this question through a three-step process
[Read: How Polestar is using blockchain to increase transparency]
First, they generated a list of drug candidates using an autoencoder, a type of neural network that finds data representations in an unsupervised way.
The autoencoder analyzed two data sets of gene expression patterns to identify drugs that appeared to neutralize the virus.
The researchers then narrowed the list down by mapping the interactions of proteins involved in aging and infection pathways. They then identified areas of overlap between the two maps.
This identified the gene expression network that a drug should target to combat COVID-19 in older patients.
A causal structure
Finally, the team used statistical algorithms to analyze causality in the network. This allowed them to identify the specific genes that a drug should target to minimize the impact of the infection.
The system highlighted the RIPK1 gene as a promising target for COVID-19 drugs. The researchers then found a list of approved drugs that act on RIPK1.
Some of them have been approved for cancer treatment, while others are already being tested in COVID-19 patients.
The researchers note that in vitro experiments and rigorous clinical trials are still needed to determine their effectiveness. But they also intend to extend their structure to other infections:
While we apply our computational platform in the context of SARS-CoV-2, our algorithms integrate data modalities that are available for many diseases, making them widely applicable.
You can read the study article at Nature Communications.
Published February 15, 2021 – 17:21 UTC