Cookies on this website
We use cookies to ensure that we give you the best experience on our website. If you click 'Continue' we'll assume that you are happy to receive all cookies and you won't see this message again. Click 'Find out more' for information on how to change your cookie settings.

Authors: Lv et al

Link to paper:

Journal/ Pre-Print: Research Square

Tags: Bioinformatics, Clinical, Gut

Research Highlights 

1. In comparison to healthy controls, the gut mycobiota of SARS-CoV-2 and H1N1 infected patients is largely characterised by the loss of various fungal species.

2. Different combinations of gut mycobiota constituents can be used to distinguish SARS-CoV-2, H1N1 and healthy patients.

3. When combining all disease states, clinical parameters, such as diarrhoea incidence and peripheral lymphocyte count correlate with certain members of the mycobiota.


This study utilises ITS sequencing and peripheral blood phenotyping of samples obtained from SARS-CoV-2, H1N1 and healthy patients to analyse the gut mycobiota in Covid-19 and H1N1 influenza. The overall species richness of the mycobiome was reduced in both H1N1 and SARS-CoV-2 patients vs healthy controls. Whilst Covid-19 patients exclusively displayed reductions in various mycobiota constituents, some elevated species were found in H1N1 vs healthy controls. Various combinations of the differentially abundant fungal species were shown to distinguish between H1N1, Covid-19 and healthy patients. When combining H1N1 and Covid-19 data, certain members of the mycobiome were found to correlate with various disease parameters.

Impact for SARS-CoV2/COVID19 research efforts

Understand the immune response to SARS-CoV2/COVID19:

Peripheral blood phenotyping includes lymphocyte and neutrophil counts alongside ELISA for pro-inflammatory cytokine

Clinical symptoms and pathogenesis of SARS-Cov2/COVID19:

Associations between clinical parameters and mycobiota aims to link mycobiota disturbances with disease state in Covid-19 and H1N1 influenza

Develop diagnostic tools for SARS-CoV2/COVID19:

Use of classifier and associations between mycobiome and clinical parameters may be considered when assessing biomarkers for Covid-19 diagnosis.

Study Type

· In silico study / bioinformatics study

· Patient Case study

Strengths and limitations of the paper


First study detailing disturbances in the gut mycobiota in Covid-19 vs H1N1 influenza and healthy controls

Standing in the field: Emerging papers have hinted to disturbances in the microbiota of SARS-COV2 positive patients (PMID: 32497191), however this is the first paper investigating specifically the intestinal mycobiota. Differences in the microbiota composition are also observed in diabetes and during aging, two patient characteristics also potentially linked to SARS-COV2 pathology.

Appropriate statistics: Lack of reported sequencing depth makes it hard to interpret results regarding reduced species richness, Shannon index, and altered mycobiota composition (see limitations). Lack of information the methods used to generate the reported ROC curves. No statistical significance applied to rank correlations associating members of the mycobiota with clinical parameters.

Viral model used:

Human cases of SARS-CoV-2 and H1N1 infection, confirmed by RT-PCR


Low. Translatability stems from i) classifier to predict SARS-CoV-2 infection and ii) correlations drawn between gut mycobiota and clinical parameters. However, i) classifier used takes into account different species to achieve different discriminations and ii) mycobiota sequencing results confounded by lack of reported depth and any associations between mycobiota and SARS-CoV-2/H1N1 further confounded by presence of underlying health conditions in patient cohorts.

Main limitations:

· Does not provide sufficient information regarding sequencing methodology, including adapter trimming, filtering and number of reads obtained pre- and post-processing. Only detected OTUs given. As sequencing depth per sample not reported, alterations in species composition may not be truly reflective of differences in disease states. Rather, changes may simply be due to differences in sequencing depths leading to greater or reduced detection of various fungal species.

· Underlying conditions, e.g. diabetes, and glucocorticoid use in Covid-19 differ to H1N1 cohort. As underlying conditions and anti-inflammatory drugs are likely to impact on gut microbiome composition, this complicates conclusions drawn when making comparisons between SARS-CoV-2, H1N1 and healthy controls. No efforts are made to account for these variables in the bioinformatic analysis, for example visualising pre-existing conditions as a parameter in PCA plots.

· Wrong figures referenced in the text – use of 2A-E over 3A-E (In the section: Fungal dysbiosis in H1N1 and Covid-19 patients)

· Classifiers used to distinguish healthy, SARS-CoV-2 and H1N1 infected individuals use different fungal species. i.e. it doesn’t appear that the authors are able to produce a single classifier which can be used to distinguish between all 3 patient groups.