Post-acute sequelae of COVID-19 (PASC) has heterogenous presentation and complex etiology. Here the authors profile peripheral blood of patients with PASC and analyze by machine-learning to identify immune and serology features that allow the stratification of PASC into inflammatory and non-inflammatory types for better diagnosis and therapy-planning.

My friend from graduate school (first author here) pulled me in to do some machine learning in a really interesting dataset.

In brief, we found that there are (at least) 2 distinct causes of what are currently lumped together as “long covid”.