The severe acute respiratory syndrome coronavirus 2 (SARS-CoV-2) emergence began in late 2019 and has since spread rapidly worldwide. The characteristics of respiratory immune response to this emerging virus is not clear. Recently, Single-cell RNA sequencing (scRNA-seq) transcriptome profiling of Bronchoalveolar lavage fluid (BALF) cells has been done to elucidate the potential mechanisms underlying in COVID-19. With the aim of better utilizing this atlas of BALF cells in response to the virus, here we propose a bioinformatics pipeline to identify candidate biomarkers of COVID-19 severity, which may help characterize BALF cells to have better mechanistic understanding of SARS-CoV-2 infection. The proposed pipeline is implemented in R and is available at https://github.com/namini94/scBALF_Hackathon.