nanoDoc: RNA modification detection using Nanopore raw reads with Deep One-Class Classification (preprint)

Advances in Nanopore single-molecule direct RNA sequencing (DRS) have presented the possibility of detecting comprehensive post-transcriptional modifications (PTMs) as an alternative to experimental approaches combined with high-throughput sequencing. It has been shown that the DRS method can detect the change in the raw electric current signal of a PTM; however, the accuracy and reliability still require improvement. Here, we presented a new software, called nanoDoc, for detecting PTMs from DRS data using a deep neural network. Current signal deviations caused by PTMs are analyzed via Deep One-Class Classification with a convolutional neural network. Using a ribosomal RNA dataset, the software archive displayed an area under the curve (AUC) accuracy of 0.96 for the detection of 23 different kinds of modifications in Escherichia coli and Saccharomyces cerevisiae. We also demonstrated a tentative classification of PTMs using unsupervised clustering. Finally, we applied this software to severe acute respiratory syndrome coronavirus 2 data and identified commonly modified sites among three groups. nanoDoc is open source (GPLv3) and available at https://github.com/uedaLabR/nanoDoc

Age-dependent regulation of SARS-CoV-2 cell entry genes and cell death programs correlates with COVID-19 disease severity (preprint)

Angiotensin-converting enzyme 2 (ACE2) maintains cardiovascular and renal homeostasis but also serves as the entry receptor for the novel severe acute respiratory syndrome coronavirus (SARS-CoV-2), the causal agent of novel coronavirus disease 2019 (COVID-19)1. COVID-19 disease severity, while highly variable, is typically lower in pediatric patients than adults (particularly the elderly), but increased rates of hospitalizations requiring intensive care are observed in infants than in older children. The reasons for these differences are unknown. To detect potential age-based correlates of disease severity, we measured ACE2 protein expression at the single cell level in human lung tissue specimens from over 100 donors from [~]4 months to 75 years of age. We found that expression of ACE2 in distal lung epithelial cells generally increases with advancing age but exhibits extreme intra- and inter-individual heterogeneity. Notably, we also detected ACE2 expression on neonatal airway epithelial cells and within the lung parenchyma. Similar patterns were found at the transcript level: ACE2 mRNA is expressed in the lung and trachea shortly after birth, downregulated during childhood, and again expressed at high levels in late adulthood in alveolar epithelial cells. Furthermore, we find that apoptosis, which is a natural host defense system against viral infection, is also dynamically regulated during lung maturation, resulting in periods of heightened apoptotic priming and dependence on pro-survival BCL-2 family proteins including MCL-1. Infection of human lung cells with SARS-CoV-2 triggers an unfolded protein stress response and upregulation of the endogenous MCL-1 inhibitor Noxa; in juveniles, MCL-1 inhibition is sufficient to trigger apoptosis in lung epithelial cells - this may limit virion production and inflammatory signaling. Overall, we identify strong and distinct correlates of COVID-19 disease severity across lifespan and advance our understanding of the regulation of ACE2 and cell death programs in the mammalian lung. Furthermore, our work provides the framework for potential translation of apoptosis modulating drugs as novel treatments for COVID-19.