Blood alternative splicing biomarker for infectious disease. Zhang et al.
Assays detecting blood transcriptome changes are being intensely studied for infectious disease diagnosis. Blood-based RNA alternative splicing events, which have not been well-characterized in pathogen infection, have potential normalization and assay platform stability advantages compared to gene expression for diagnosis. Here, we present a computational framework for developing robust alternative splicing diagnostic biomarkers. The framework analyzes patient whole-blood RNA sequencing (RNA-seq) data to identify and optimize alternative splicing biomarkers for diagnosing infectious disease. Leveraging a large prospective cohort of SARS-CoV-2 infection, we identify a major splicing program switch upon viral infection. Functional analysis reveals significant enrichment of differential splicing events in immune-specific protein domains and genes. Using test whole-blood RNA-seq data from an independent cohort, we demonstrate the superiority of alternative splicing biomarkers for SARS-CoV-2 diagnosis compared to six reported transcriptome signatures. We then optimize a subset of alternative splicing-based biomarkers to develop microfluidic PCR diagnostic assays for SARS-CoV-2 diagnosis. This assay achieves nearly perfect test accuracy (61/62=98.4%; positive percent agreement=96.7%, negative percent agreement=100%) using a naive principal component classifier, which was significantly more accurate than a gene expression PCR assay tested in the same cohort. Therefore, our RNA splicing computational framework enables a promising avenue for host response diagnosis of infection.
Steps to reproduce
Please use the Jupyter notebooks here for reproducing these results: https://github.com/zj-zhang/CHARM-AlternativeSplicing