A machine-learning approach to finding treatment options for Covid-19

When the Covid-19 pandemic struck in soon 2020 doctors and investigationers rushed to find powerful handlements. There was pliant time to spare. ’Making new drugs takes forever’ says Caroline Uhler a computational biologist in MITs Department of Electrical Engineering and Computer Science. ’Really the only expedient discretion is to repurpose existing drugs.’

Uhlers team has now developed a machine learning-based access to unite drugs already on the market that could possiblely be repurposed to battle Covid-19 specially in the elderly. The method accounts for changes in gene countenance in lung cells caused by both the disease and aging. That union could allow medical experts to more quickly seek drugs for clinical testing in elderly patients who tend to experience more serious symptoms. The investigationers pinpointed the protein RIPK1 as a promising target for Covid-19 drugs and they identified three approved drugs that act on the countenance of RIPK1.

The investigation appears today in the journal Nature Communications. Co-authors include MIT PhD students Anastasiya Belyaeva Adityanarayanan Radhakrishnan Chandler Squires and Karren Dai Yang as well as PhD student Louis Cammarata of Harvard University and long-term collaborator G.V. Shivashankar of ETH Zurich in Switzerland. Uhler is also an companion limb of the Broad Institute of MIT and Harvard and a investigationer at MITs Laboratory for Information amp; Decision Systems (LIDS) and the Institute for Data Systems and Society (IDSS).

Early in the pandemic it grew clear that Covid-19 harmed older patients more than younger ones on mean. Uhlers team wondered why. ’The prevalent hypothesis is the aging immune method’ she says. But Uhler and Shivashankar suggested an additional factor: ’One of the main changes in the lung that happens through aging is that it becomes stiffer.’

The stiffening lung tissue shows different patterns of gene countenance than in younger nation even in response to the same eminent. ’Earlier work by the Shivashankar lab showed that if you stimulate cells on a stiffer substrate with a cytokine correspondent to what the virus does they verity turn on different genes’ says Uhler. ’So that motivated this hypothesis. We need to look at aging unitedly with SARS-CoV-2 — what are the genes at the intersection of these two pathways?’ To select approved drugs that might act on these pathways the team turned to big data and artificial intelligence.

The investigationers zeroed in on the most promising drug repurposing aspirants in three wide steps. First they engenderd a big list of practicable drugs using a machine-learning technique named an autoencoder. Next they mapped the network of genes and proteins implicated in both aging and SARS-CoV-2 taint. Finally they used statistical algorithms to apprehend eventuality in that network allowing them to pinpoint ’upstream’ genes that caused cascading effects throughout the network. In source drugs targeting those upstream genes and proteins should be promising aspirants for clinical trials.

To engender an initial list of possible drugs the teams autoencoder relied on two key datasets of gene countenance patterns. One dataset showed how countenance in different cell types responded to a range of drugs already on the market and the other showed how countenance responded to taint with SARS-CoV-2. The autoencoder scoured the datasets to highlight drugs whose contacts on gene countenance appeared to counterinfluence the effects of SARS-CoV-2. ’This application of autoencoders was challenging and required foundational insights into the working of these neural networks which we developed in a paper recently published in PNAS’ notes Radhakrishnan.

Next the investigationers narrowed the list of possible drugs by homing in on key genetic pathways. They mapped the interactions of proteins implicated in the aging and Sars-CoV-2 taint pathways. Then they identified areas of overlap among the two maps. That effort pinpointed the definite gene countenance network that a drug would need to target to combat Covid-19 in elderly patients.

’At this point we had an undirected network’ says Belyaeva signification the investigationers had yet to unite which genes and proteins were ’upstream’ (i.e. they have cascading effects on the countenance of other genes) and which were ’downstream’ (i.e. their countenance is altered by preceding changes in the network). An mental drug aspirant would target the genes at the upstream end of the network to minimize the contacts of taint.

’We want to unite a drug that has an effect on all of these differentially expressed genes downstream’ says Belyaeva. So the team used algorithms that gather eventuality in interacting methods to turn their undirected network into a causal network. The terminal causal network identified RIPK1 as a target gene/protein for possible Covid-19 drugs since it has numerous downstream effects. The investigationers identified a list of the approved drugs that act on RIPK1 and may have possible to handle Covid-19. Previously these drugs have been approved for the use in cancer. Other drugs that were also identified including ribavirin and quinapril are already in clinical trials for Covid-19.

Uhler plans to share the teams findings with pharmaceutical companies. She emphasizes that precedently any of the drugs they identified can be approved for repurposed use in elderly Covid-19 patients clinical testing is needed to determine efficiency. While this particular study focused on Covid-19 the investigationers say their framework is extendable. ’Im veritably excited that this platform can be more generally applied to other taints or diseases’ says Belyaeva. Radhakrishnan emphasizes the weight of gathering information on how different diseases contact gene countenance. ’The more data we have in this space the better this could work’ he says.

This investigation was supported in part by the Office of Naval Research the National Science Foundation the Simons Foundation the MIT-IBM Watson AI Lab and the MIT Jameel Clinic for Machine Learning and Health.