An ensemble prediction model for COVID-19 mortality risk.

Background: It's critical to identify COVID-19 patients with a higher death risk at early stage to give them better hospitalization or intensive care. However, thus far, none of the machine learning models has been shown to be successful in an independent cohort. We aim to develop a machine learning model which could accurately predict death risk of COVID-19 patients at an early stage in other independent cohorts. Methods: We used a cohort containing 4711 patients whose clinical features associated with patient physiological conditions or lab test data associated with inflammation, hepatorenal function, cardiovascular function, and so on to identify key features. To do so, we first developed a novel data preprocessing approach to clean up clinical features and then developed an ensemble machine learning method to identify key features. Results: Finally, we identified 14 key clinical features whose combination reached a good predictive performance of area under the receiver operating characteristic curve 0.907. Most importantly, we successfully validated these key features in a large independent cohort containing 15 790 patients. Conclusions: Our study shows that 14 key features are robust and useful in predicting the risk of death in patients confirmed SARS-CoV-2 infection at an early stage, and potentially useful in clinical settings to help in making clinical decisions.

Functional and genetic analysis of viral receptor ACE2 orthologs reveals a broad potential host range of SARS-CoV-2.

The pandemic of COVID-19, caused by SARS-CoV-2, is a major global health threat. Epidemiological studies suggest that bats (Rhinolophus affinis) are the natural zoonotic reservoir for SARS-CoV-2. However, the host range of SARS-CoV-2 and intermediate hosts that facilitate its transmission to humans remain unknown. The interaction of coronavirus with its host receptor is a key genetic determinant of host range and cross-species transmission. SARS-CoV-2 uses angiotensin-converting enzyme 2 (ACE2) as the receptor to enter host cells in a species-dependent manner. In this study, we characterized the ability of ACE2 from diverse species to support viral entry. By analyzing the conservation of five residues in two virus-binding hotspots of ACE2 (hotspot 31Lys and hotspot 353Lys), we predicted 80 ACE2 proteins from mammals that could potentially mediate SARS-CoV-2 entry. We chose 48 ACE2 orthologs among them for functional analysis, and showed that 44 of these orthologs-including domestic animals, pets, livestock, and animals commonly found in zoos and aquaria-could bind the SARS-CoV-2 spike protein and support viral entry. In contrast, New World monkey ACE2 orthologs could not bind the SARS-CoV-2 spike protein and support viral entry. We further identified the genetic determinant of New World monkey ACE2 that restricts viral entry using genetic and functional analyses. These findings highlight a potentially broad host tropism of SARS-CoV-2 and suggest that SARS-CoV-2 might be distributed much more widely than previously recognized, underscoring the necessity to monitor susceptible hosts to prevent future outbreaks.

Clinical sequelae of COVID-19 survivors in Wuhan, China: a single-centre longitudinal study.

OBJECTIVES: To describe the prevalence, nature and risk factors for the main clinical sequelae in coronavirus disease 2019 (COVID-19) survivors who have been discharged from the hospital for more than 3 months. METHODS: This longitudinal study was based on a telephone follow-up survey of COVID-19 patients hospitalized and discharged from Renmin Hospital of Wuhan University, Wuhan, China before 1 March 2020. Demographic and clinical characteristics and self-reported clinical sequelae of the survivors were described and analysed. A cohort of volunteers who were free of COVID-19 and lived in the urban area of Wuhan during the outbreak were also selected as the comparison group. RESULTS: Among 538 survivors (293, 54.5% female), the median (interquartile range) age was 52.0 (41.0-62.0) years, and the time from discharge from hospital to first follow-up was 97.0 (95.0-102.0) days. Clinical sequelae were common, including general symptoms (n = 267, 49.6%), respiratory symptoms (n = 210, 39%), cardiovascular-related symptoms (n = 70, 13%), psychosocial symptoms (n = 122, 22.7%) and alopecia (n = 154, 28.6%). We found that physical decline/fatigue (p < 0.01), postactivity polypnoea (p= 0.04) and alopecia (p < 0.01) were more common in female than in male subjects. Dyspnoea during hospitalization was associated with subsequent physical decline/fatigue, postactivity polypnoea and resting heart rate increases but not specifically with alopecia. A history of asthma during hospitalization was associated with subsequent postactivity polypnoea sequela. A history of pulse ≥90 bpm during hospitalization was associated with resting heart rate increase in convalescence. The duration of virus shedding after COVID-19 onset and hospital length of stay were longer in survivors with physical decline/fatigue or postactivity polypnoea than in those without. CONCLUSIONS: Clinical sequelae during early COVID-19 convalescence were common; some of these sequelae might be related to gender, age and clinical characteristics during hospitalization.