Knowledge synthesis from 100 million biomedical documents augments the deep expression profiling of coronavirus receptors

AJ Venkatakrishnan; Arjun Puranik; Akash Anand; David Zemmour; Xiang Yao; Xiaoying Wu; Ramakrishna Chilaka; Dariusz K Murakowski; Kristopher Standish; Bharathwaj Raghunathan; Tyler Wagner; Enrique Garcia-Rivera; Hugo Solomon; Abhinav Garg; Rakesh Barve; Anuli Anyanwu-Ofili; Najat Khan; Venky Soundararajan

To determine whether the maturation-correlated expression pattern is unique to ACE2, we computed cosine similarities between the ACE2 gene expression vector (CP10K values in ~6,000 small intestinal enterocytes) and that of the ~15,700 other genes detected in this study ( Figure 5A ). For this analysis, the vector space is constituted of the individual cells as the dimensions using the gene expression values to construct the vectors (see Methods). Interestingly, we found that ANPEP, the established entry receptor for HCoV-229E, showed the third highest cosine similarity to ACE2 ( Figure 5B ). Further, DPP4 -the entry receptor for MERS coronavirus -is also among the top 1% of similarly expressed genes by this metric (Figure 5B ). We confirmed that both of these genes do indeed show a maturation-correlated transcriptional pattern similar to that of ACE2 ( Figure 5C-D) , highlighting an unexpected shared pattern of transcriptional heterogeneity among known coronavirus receptors in a cell population which may be relevant for viral transmission.