Disease diagnostics using machine learning of immune receptors (preprint)

ABSTRACT Clinical diagnoses rely on a wide variety of laboratory tests and imaging studies, interpreted alongside physical examination and documentation of symptoms and patient history. However, the tools of diagnosis make little use of the immune system’s internal record of specific disease exposures encoded by the antigen-specific receptors of memory B cells and T cells. We have combined extensive receptor sequence datasets with three different machine learning representations of the contents of immune repertoires to develop an interpretive framework, MAchine Learning for Immunological Diagnosis (Mal-ID) , that screens for multiple illnesses simultaneously. This approach can already reliably distinguish a wide range of disease states, including specific acute or chronic infections, and autoimmune or immunodeficiency disorders, and could contribute to identifying new infectious diseases as they emerge. Importantly, many features of the model of immune receptor sequences are human-interpretable. They independently recapitulate known biology of the responses to infection by SARS-CoV-2 or HIV, and reveal common features of autoreactive immune receptor repertoires, indicating that machine learning on immune repertoires can yield new immunological knowledge.

A serological assay to detect SARS-CoV-2 seroconversion in humans (preprint)

SARS-Cov-2 (severe acute respiratory disease coronavirus 2), which causes Coronavirus Disease 2019 (COVID19) was first detected in China in late 2019 and has since then caused a global pandemic. While molecular assays to directly detect the viral genetic material are available for the diagnosis of acute infection, we currently lack serological assays suitable to specifically detect SARS-CoV-2 antibodies. Here we describe serological enzyme-linked immunosorbent assays (ELISA) that we developed using recombinant antigens derived from the spike protein of SARS-CoV-2. Using negative control samples representing pre-COVID 19 background immunity in the general adult population as well as samples from COVID19 patients, we demonstrate that these assays are sensitive and specific, allowing for screening and identification of COVID19 seroconverters using human plasma/serum as early as two days post COVID19 symptoms onset. Importantly, these assays do not require handling of infectious virus, can be adjusted to detect different antibody types and are amendable to scaling. Such serological assays are of critical importance to determine seroprevalence in a given population, define previous exposure and identify highly reactive human donors for the generation of convalescent serum as therapeutic. Sensitive and specific identification of coronavirus SARS-Cov-2 antibody titers may, in the future, also support screening of health care workers to identify those who are already immune and can be deployed to care for infected patients minimizing the risk of viral spread to colleagues and other patients.