Common, low-frequency, rare, and ultra-rare coding variants contribute to COVID-19 severity. (preprint)

The combined impact of common and rare exonic variants in COVID-19 host genetics is currently insufficiently understood. Here, common and rare variants from whole exome sequencing data of about 4,000 SARS-CoV-2-positive individuals were used to define an interpretable machine learning model for predicting COVID-19 severity. Firstly, variants were converted into separate sets of Boolean features, depending on the absence or the presence of variants in each gene. An ensemble of LASSO logistic regression models was used to identify the most informative Boolean features with respect to the genetic bases of severity. The Boolean features selected by these logistic models were combined into an Integrated PolyGenic Score that offers a synthetic and interpretable index for describing the contribution of host genetics in COVID-19 severity, as demonstrated through testing in several independent cohorts. Selected features belong to ultra-rare, rare, low-frequency, and common variants, including those in linkage disequilibrium with known GWAS loci. Noteworthly, around one quarter of the selected genes are sex-specific. Pathway analysis of the selected genes associated with COVID-19 severity reflected the multi-organ nature of the disease. The proposed model might provide useful information for developing diagnostics and therapeutics, while also being able to guide bedside disease management.

Cryptic SARS-CoV2-spike-with-sugar interactions revealed by 'universal' saturation transfer analysis (preprint)

Host-expressed proteins on both host-cell and pathogen surfaces are widely exploited by pathogens, mediating cell entry (and exit) and influencing disease progression and transmission. This is highlighted by the diverse modes of coronavirus entry into cells and their consequent differing pathogenicity that is of direct relevance to the current SARS-CoV-2 pandemic. Host-expressed viral surface proteins bear post-translational modifications such as glycosylation that are essential for function but can confound or limit certain current biophysical methods used for dissecting key interactions. Several human coronaviruses attach to host cell-surface N-linked glycans that include forms of sialic acid. There remains, however, conflicting evidence as to if or how SARS-associated coronaviruses might use such a mechanism. Here, we show that novel protein NMR methods allow a complete and comprehensive analysis of the magnetization transfer caused by interactions between even heavily modified proteins and relevant ligands to generate quantitative binding data in a general manner. Our method couples direct, objective resonance-identification via a deconvolution algorithm with quantitative analysis using Bloch-McConnell equations to obtain interaction parameters (e.g. KD, kEx), which together enable structural modelling. By using an automated and openly available workflow, this method can be readily applied in a range of systems. This complete treatment of so-called 'saturation transfer' between protein and ligand now enables a general analysis of solution-phase ligand-protein binding beyond previously perceived limits of exchange rates, concentration or system - this allows 'universal' saturation transfer analysis (uSTA). uSTA proves critical in mapping direct interaction between natural sialoside sugar ligands and SARS-CoV-2-spike glycoprotein by quantitating ligand signal in spectral regions otherwise occluded by resonances from mobile spike-protein glycans (that also include sialosides). Using uSTA, 'end on'-binding by SARS-CoV-2-spike protein to sialoside glycan is revealed, which contrasts with an observed 'extended surface'-binding for previously validated heparin sugar ligands. Quantitative use of uSTA-derived restraints pinpoints likely binding modes to an intrinsically disordered region of the N-terminal domain of SARS-CoV-2-spike trimer. Consistent with this, glycan binding is minimally perturbed by antibodies that neutralize via binding the ACE2-binding domain (RBD) but strongly disrupted in the B1.1.7 and B1.351 variants-of-concern that possess hotspot mutations around the identified site. An analysis of beneficial genetic variances in cohorts of patients from early 2020 suggests a possible model in which A-lineage-SARS-CoV-2 may have exploited a specific sialylated-polylactosamine motif found on tetraantennary human N-linked-glycoproteins in deeper lung. Since cell-surface glycans are widely relevant to biology and pathology, uSTA can now provide a ready, quantitative method for widespread analysis of complex, host-derived and post-translationally modified proteins with putative ligands relevant to disease even in previously confounding complex systems.

Post-Mendelian genetic model in COVID-19 (preprint)

Host genetics is an emerging theme in COVID-19 and few common polymorphisms and some rare variants have been identified, either by GWAS or candidate gene approach, respectively. However, an organic model is still missing. Here, we propose a new model that takes into account common and rare germline variants applied in a cohort of 1,300 Italian SARS-CoV-2 positive individuals. Ordered logistic regression of clinical WHO grading on sex and age was used to obtain a binary phenotypic classification. Genetic variability from WES was synthesized in several boolean representations differentiated according to allele frequencies and genotype effect. LASSO logistic regression was used for extracting relevant genes. We defined about 100 common driver polymorphisms corresponding to classical "threshold model". Extracted genes were demonstrated to be gender specific. Stochastic rare more penetrant events on about additional 100 extracted genes, when occurred in a medium or severe background (common within the family), simulate Mendelian inheritance in 14% of subjects (having only 1 mutation) or oligogenic inheritance (in 10% having 2 mutations, in 11% having 3 mutations, etc). The combined effect of common and rare results can be described as an integrated polygenic score computed as: (nseverity - nmildness) + F (mseverity - mmildness) where n is the number of common driver genes, m is the number of driver rare variants and F is a factor for appropriately weighing the more powerful rare variants. We called the model "post-Mendelian". The model well describes the cohort, and patients are clustered in severe or mild by the integrated polygenic scores, the F factor being calibrated around 2, with a prediction capacity of 65% in males and 70% in females. In conclusion, this is the first comprehensive model interpreting host genetics in a holistic post-Mendelian manner. Further validations are needed in order to consolidate and refine the model which however holds true in thousands of SARS-CoV-2 Italian subjects.

SARS-CoV-2 Seroprevalence Survey in People Involved in Different Essential Activities during the General Lock-Down Phase in the Province of Prato (Tuscany, Italy)

Serosurveys may help to assess the transmission dynamics in high-risk groups. The aim of the study was to assess the SARS-CoV-2 antibody seroprevalence in people who had performed essential activities during the lock-down period in the Province of Prato (Italy), and to evaluate the risk of exposure to SARS-CoV-2 according to the type of service. All the workers and volunteers of the Civil Protection, employees of the municipalities, and all the staff of the Health Authority of the Province of Prato were invited to be tested with a rapid serological test. A total of 4656 participants were tested. SARS-CoV-2 antibodies were found in 138 (2.96%) cases. The seroprevalence in health care workers, in participants involved in essential support services and in those who worked from home were 4.1%, 1.4% and 1.0%, respectively. Health care workers experienced higher odds of seropositivity (OR 4.38, 95%CI 2.19–10.41) than participants who were assigned to work-from-home;no significant seropositivity differences were observed between support services and work-from-home groups. A low circulation of SARS-CoV-2 was observed among participants performing different essential activities. Findings highlighted the risk of in-hospital transmission in healthcare workers and that community support services may increase the risk of seropositivity to a limited extent in low incidence areas.

Shorter androgen receptor polyQ alleles protect against life-threatening COVID-19 disease in males. (preprint)

BackgroundCOVID-19 presentation ranges from asymptomatic to fatal. The variability in severity may be due in part to impaired Interferon type I response due to specific mutations in the host genome or to autoantibodies, explaining about 15% of the cases when combined. Exploring the host genome is thus warranted to further elucidate disease variability. MethodsWe developed a synthetic approach to genetic data representation using machine learning methods to investigate complementary genetic variability in COVID-19 infected patients that may explain disease severity, due to poly-amino acids repeat polymorphisms. Using host whole-exome sequencing data, we compared extreme phenotypic presentations (338 severe versus 300 asymptomatic cases) of the entire (men and women) Italian GEN-COVID cohort of 1178 subjects infected with SARS-CoV-2. We then applied the LASSO Logistic Regression model on Boolean gene-based representation of the poly-amino acids variability. FindingsShorter polyQ alleles ([≤]22) in the androgen receptor (AR) conferred protection against a more severe outcome in COVID-19 infection. In the subgroup of males with age <60 years, testosterone was higher in subjects with AR long-polyQ ([≥]23), possibly indicating receptor resistance (p=0.004 Mann-Whitney U test). Inappropriately low testosterone levels for the long-polyQ alleles predicted the need for intensive care in COVID-19 infected men. In agreement with the known anti-inflammatory action of testosterone, patients with long-polyQ ([≥]23) and age>60 years had increased levels of C Reactive Protein (p=0.018). InterpretationOur results may contribute to design reliable clinical and public health measures and provide a rationale to test testosterone treatment as adjuvant therapy in symptomatic COVID-19 men expressing AR polyQ longer than 23 repeats. FundingMIUR project "Dipartimenti di Eccellenza 2018-2020" to Department of Medical Biotechnologies University of Siena, Italy (Italian D.L. n.18 March 17, 2020). Private donors for COVID research and charity funds from Intesa San Paolo. BoxesO_ST_ABSEvidence before this studyC_ST_ABSWe searched on Medline, EMBASE, and Pubmed for articles published from January 2020 to August 2020 using various combinations of the search terms "sex-difference", "gender" AND SARS-Cov-2, or COVID. Epidemiological studies indicate that men and women are similarly infected by COVID-19, but the outcome is less favorable in men, independently of age. Several studies also showed that patients with hypogonadism tend to be more severely affected. A prompt intervention directed toward the most fragile subjects with SARS-Cov2 infection is currently the only strategy to reduce mortality. glucocorticoid treatment has been found cost-effective in improving the outcome of severe cases. Clinical algorithms have been proposed, but little is known on the ability of genetic profiling to predict outcome and disclose novel therapeutic strategies. Added-value of this studyIn a cohort of 1178 men and women with COVID-19, we used a supervised machine learning approach on a synthetic representation of the uncovered variability of the human genome due to poly-amino acid repeats. Comparing the genotype of patients with extreme manifestations (severe vs. asymptomatic), we found that the poly-glutamine repeat of the androgen receptor (AR) gene is relevant for COVID-19 disease and defective AR signaling identifies an association between male sex, testosterone exposure, and COVID-19 outcome. Failure of the endocrine feedback to overcome AR signaling defect by increasing testosterone levels during the infection leads to the fact that polyQ becomes dominant to T levels for the clinical outcome. Implications of all the available evidenceWe identify the first genetic polymorphism predisposing some men to develop a more severe disease irrespectively of age. Based on this, we suggest that sizing the AR poly-glutamine repeat has important implications in the diagnostic pipeline of patients affected by life-threatening COVID-19 infection. Most importantly, our studies open to the potential of using testosterone as adjuvant therapy for severe COVID-19 patients having defective androgen signaling, defined by this study as [≥]23 PolyQ repeats and inappropriate levels of circulating androgens.

Shorter Androgen Receptor PolyQ Alleles Protect Against Life-Threatening COVID-19 Disease in Males (preprint)

Background: COVID-19 presentation ranges from asymptomatic to fatal. The variability in severity is due in part to specific mutations in the host genome. GWAS effectively identifies genetic variability due to common biallelic polymorphisms. Efforts in genetic research are trying to identify significant associations in patients infected by SARS-CoV-2. Methods: We developed a synthetic approach to genetic data representation using machine learning methods to investigate complementary genetic variability in COVID-19 infected patients that might explain disease severity due to rare variants and poly-amino acids repeat polymorphisms. Using host whole-exome sequencing data, we compared extreme phenotypic presentations of an Italian cohort of 939 subjects infected with SARS-CoV-2. We then applied the LASSO Logistic Regression model on Boolean gene-based representation of the entire set of human genes. Findings: Polymorphisms/rare variants in certain genes, including short polyQ (≤22) of the androgen receptor ( AR ), conferred protection against severe forms of COVID-19. We then demonstrated that testosterone was higher in males with AR long-polyQ (≥23), confirming receptor resistance (p=0.004 Mann-Whitney U test). Finally, long-polyQ (≥23) correlates with increased inflammation markers (p=0.021) and fibrinogen consumption (p=0.039), confirming the anti-inflammatory role of testosterone. Interpretation: Our results contribute to designing reliable clinical and public health measures and provide a rationale to test testosterone supplementation as adjuvant treatment in symptomatic COVID-19 men expressing AR polyQ longer than 23. Funding: MIUR project “Dipartimenti di Eccellenza 2018-2020” to Department of Medical Biotechnologies University of Siena, Italy; Private donors for COVID research (Italian D.L. n.18 March 17, 2020).Declaration of Interests: The authors declare no competing interests.Ethics Approval Statement: The GEN-COVID study was approved by the University Hospital of Siena Ethics Review Board (Protocol n. 16917, dated March 16, 2020).

An Optimal Control Approach to Learning in SIDARTHE Epidemic model (preprint)

The COVID-19 outbreak has stimulated the interest in the proposal of novel epidemiological models to predict the course of the epidemic so as to help planning effective control strategies. In particular, in order to properly interpret the available data, it has become clear that one must go beyond most classic epidemiological models and consider models that, like the recently proposed SIDARTHE, offer a richer description of the stages of infection. The problem of learning the parameters of these models is of crucial importance especially when assuming that they are time-variant, which further enriches their effectiveness. In this paper we propose a general approach for learning time-variant parameters of dynamic compartmental models from epidemic data. We formulate the problem in terms of a functional risk that depends on the learning variables through the solutions of a dynamic system. The resulting variational problem is then solved by using a gradient flow on a suitable, regularized functional. We forecast the epidemic evolution in Italy and France. Results indicate that the model provides reliable and challenging predictions over all available data as well as the fundamental role of the chosen strategy on the time-variant parameters.

Employing a Systematic Approach to Biobanking and Analyzing Genetic and Clinical Data for Advancing COVID-19 Research (preprint)

Within the GEN-COVID Multicenter Study, biospecimens from more than 1,000 SARS-CoV-2-positive individuals have thus far been collected in the GEN-COVID Biobank (GCB). Sample types include whole blood, plasma, serum, leukocytes, and DNA. The GCB links samples to detailed clinical data available in the GEN-COVID Patient Registry (GCPR). It includes hospitalized patients (74.25%), broken down into intubated, treated by CPAP-biPAP, treated with O2 supplementation, and without respiratory support (9.5%, 18.4%, 31.55% and 14.8, respectively); and non-hospitalized subjects (25.75%), either pauci- or asymptomatic. More than 150 clinical patient-level data fields have been collected and binarized for further statistics according to the organs/systems primarily affected by COVID-19: heart, liver, pancreas, kidney, chemosensors, innate or adaptive immunity, and clotting system. Hierarchical Clustering analysis identified five main clinical categories: i) severe multisystemic failure with either thromboembolic or pancreatic variant; ii) cytokine storm type, either severe with liver involvement or moderate; iii) moderate heart type, either with or without liver damage; iv) moderate multisystemic involvement, either with or without liver damage; v) mild, either with or without hyposmia. GCB and GCPR are further linked to the GEN-COVID Genetic Data Repository (GCGDR), which includes data from Whole Exome Sequencing and high-density SNP genotyping. The data are available for sharing through the Network for Italian Genomes, found within the COVID-19 dedicated section. The study objective is to systematize this comprehensive data collection and begin identifying multi-organ involvement in COVID-19, defining genetic parameters for infection susceptibility within the population and mapping genetically COVID-19 severity and clinical complexity among patients.