Prediction of severe infections based on SARS-CoV-2 genomic analysis

A group of scientists claim that the severity of coronavirus disease 2019 (COVID-19) depends on the method of exposure, the pathogenicity of the causative agent, the susceptibility of the host and its response to the pathogen.

study: SARS-CoV-2 genome-based severity prediction corresponds to lower qPCR values ​​and higher viral load.. Image Credit: Adao / Shutterstock

However, the current pandemic of COVID-19, caused by the rapid outbreak of severe acute respiratory syndrome coronavirus-2 (SARS-CoV-2), shows considerable diversity in both the host and the virus and is broad clinical. Shows the outcome.

Background

Previously, disease severity was primarily associated with host phenotypes such as gender, age, and blood type. This pandemic shows that geographic regions, viral mutations, and host genetic susceptibility play important roles in serious clinical outcomes.

To predict the severity of the illness an individual experiences, scientists used computational models based on phenotypic, genetic, and demographic data.

The main purpose of these models is to tailor the optimal treatment for patients infected with SARS-CoV-2. Early prediction of the severity of the disease can help maintain life and health.

Scientists said real-time PCR data with increased cycle thresholds could be associated with a 9% reduction in in-hospital mortality. They further found that patients with a cycle threshold below 23 were 3.9 times more likely to die in the hospital than patients with a cycle threshold above 33.

Previous studies have shown that the use of PCR cycle thresholds is an effective predictor of COVID-19 results, but it is indistinguishable from different levels of disease severity.

Scientists have used the SARS-CoV-2 genome-wide sequence to identify newly emerging mutants. Some mutants are classified as of concern because of their high pathogenicity, infectivity, and avoidance of immune response by vaccines or natural infections. These variants have emerged due to mutations in spike sequences and other parts of the viral genome.

Using these SARS-CoV-2 sequence data, a group of researchers have developed an algorithm that predicts severity based on viral mutations.

They also identified 17 mutants associated with severe clinical outcomes, with 67 mutants associated with mild clinical manifestations. This report showed the ability to discriminate to classify critically ill patients by performing areas under curve analysis.Researchers have recently focused on assessing whether genome-based prediction algorithms developed to predict clinical severity can also predict the results of the polymerase chain reaction (PCR) on behalf of decisions. Viral load And severity.

This study is available at medRxiv* Preprint server.

About research

Current research comprehensively validates predictions from machine learning models and establishes their reliability through powerful analytical tools for predicting disease severity. Scientists are convinced that their results will help clinicians determine treatment lines tailored to a particular patient. The current study used an outgroup sample containing orthogonal severity markers that supported the algorithm. This allows you to identify strains of virus that are biologically unique and have significant clinical differences.

Previous studies related to Middle East Respiratory Syndrome (MERS) virus have reported that viral load is strongly associated with severe infection and mortality. The study also stated that lowering the cycle threshold increased the risk of death by 17%.

However, this study could not include factors such as age, viral load, and disease severity correlations. Currently, an individual is considered “extremely infectious” if the cycle threshold is less than 20.

Researchers used a previously published algorithm that was used to compare viral genome-based severity predictions to the PCR-based viral load of clinically derived 716 viral genomes. The average cycle threshold for samples representing severe COVID-19 results was 18.3. In addition, the average cycle threshold for people with mild symptoms was 20.4. Current studies predict that there is a significant correlation between the predicted severity probability and the cycle threshold.

Conclusion

This study had several limitations, including the use of PCR cycle thresholds as a surrogate for clinical severity. However, PCR testing can only fairly predict clinical outcomes and is not an ideal measure.

Another limitation of this study was the low number of viral genotypes.

Despite these limitations, this study showed that genome-based algorithms can be linked to criteria for clinical diagnostic tests that can predict the severity of COVID-19.

Researchers have shown that genetic information about the virus and patient demographics may help clinicians determine the appropriate COVID-19 treatment for infected individuals.

In addition, with SARS-CoV-2 sequence data In silicoDerived severity markers may help in the design of new mutant vaccines.

The study also said that genomic surveillance can help identify new strains of the virus that may be endemic, thereby providing health authorities with sufficient time to prepare strategies to contain the infection.

*Important Notices

medRxiv publishes unpeer-reviewed preliminary scientific reports and should not be considered definitive, guide clinical / health-related behaviors, or be treated as established information.