Skip to main content

Computational network analysis of host genetic risk variants of severe COVID-19



Genome-wide association studies have identified numerous human host genetic risk variants that play a substantial role in the host immune response to SARS-CoV-2. Although these genetic risk variants significantly increase the severity of COVID-19, their influence on body systems is poorly understood. Therefore, we aim to interpret the biological mechanisms and pathways associated with the genetic risk factors and immune responses in severe COVID-19. We perform a deep analysis of previously identified risk variants and infer the hidden interactions between their molecular networks through disease mapping and the similarity of the molecular functions between constructed networks.


We designed a four-stage computational workflow for systematic genetic analysis of the risk variants. We integrated the molecular profiles of the risk factors with associated diseases, then constructed protein–protein interaction networks. We identified 24 protein–protein interaction networks with 939 interactions derived from 109 filtered risk variants in 60 risk genes and 56 proteins. The majority of molecular functions, interactions and pathways are involved in immune responses; several interactions and pathways are related to the metabolic and cardiovascular systems, which could lead to multi-organ complications and dysfunction.


This study highlights the importance of analyzing molecular interactions and pathways to understand the heterogeneous susceptibility of the host immune response to SARS-CoV-2. We propose new insights into pathogenicity analysis of infections by including genetic risk information as essential factors to predict future complications during and after infection. This approach may assist more precise clinical decisions and accurate treatment plans to reduce COVID-19 complications.


Immune-mediated inflammatory lung damage is caused by severe COVID-19 infections. Following SARS-CoV-2 infection, host genetic variations contribute to the development of illnesses that require critical care or hospitalization [1, 2]. Genome-wide association studies (GWAS) have been conducted to identify and validate risk genetic variants that significantly impact the severity of COVID-19 among different populations [3, 4]. These studies compared the genomes of patients with severe infections to those of uninfected or mildly affected individuals in order to understand the heterogeneity of the immune responses observed among patients with COVID-19 and discover the underlying disease mechanisms. Although GWAS can be leveraged to determine the associations between genetic variants and the severity of COVID-19, most of the risk variants identified are located in non-coding loci. Some of these studies applied statistical analyses such as Mendelian Randomization (MR) [5] and co-localization [6]. Such approaches help in identifying novel genetic variants in various loci in the host human genome that are associated with the severity of COVID-19 and respiratory failure [6, 7]. Although non-coding variants cannot be directly interpreted, these loci often surround multiple genes. Hence, researchers can infer the effects of these loci on gene expression by integrating multi-omics data [8].

Although previous studies have identified risk variants associated with severe COVID-19, the biological functions and pathways of most identified risk variants and their impacts on the immune system are poorly explained [9, 10]. Such pathways can be interpreted by constructing molecular networks for omics identified as risk factors for severe COVID-19. For instance, various immunological studies concentrated on examining the susceptibility of the immune system response to SARS-Cov-2. Two studies have examined the effects of variants in developing severe COVID-19 by analyzing the function of active proteins of Cytokines response and identifying genetic factors in the interferon circuit during infection [10, 11]. The studies have demonstrated that interferons (IFN-I) play vital roles in the governance of the pathogenesis of COVID-19. The genetic analysis that has been found in the enrichment of (IFN-I) genes and autoantibodies assists doctors in stating and determining the individuals who are at the critical stage of life-threatening COVID-19 [10, 12, 13].Furthermore, functional enrichment analysis can be applied to estimate the molecular function of each risk factor and its contribution to the constructed network. However, there has been no combined analysis of the interactions between all of the proteins in these networks; thus, the interactions or links between these risk factors remain unknown. Hence, mapping genetic risk variants with related diseases and biomarkers is vital to discover and predict possible missing interactions and expand the networks [14].

Thus, in this paper, we aim to further interpret the biological mechanisms and molecular pathways involved in the pathogenicity of severe COVID-19 by analyzing identified risk variants and inferring the hidden interactions between these molecular networks based on disease mapping. A computational analysis workflow is designed to retrieve genetic risk variants from several biological public resources and available platforms and annotate these risk variants with features to provide deeper insight into each risk variant. We adopted a disease-variant-network mapping approach to perform enrichment analysis and identify the similarity between molecular functions and networks, connect the constructed networks, and infer the invisible links between the constructed protein–protein interaction (PPI) networks. Finally, we map the genetic features and shared biological pathways of the phenotype of the host response to SARS-CoV-2. This analysis identified relevant host genetic factors that are involved in the progression of severe COVID-19 and also related to other diseases, such as metabolic diseases. Thus, this work could help clinicians to identify the hidden genetic factors that influence severe outcomes before symptoms appear and provide more intensive or timely treatment to reduce disease complications. A general overview of the study is visualized in the following flow diagram to illustrate our analysis concept is shown in Fig. 1.

Fig. 1
figure 1

Flow diagram of the study overview. (A) The first step: extracting genetic risk factors of severe COVID-19, by parsing articles and extracting risk factors. such as Variants, Genes, proteins, related diseases, pathways, and interactions. (B) The second step: annotating the genetic risk factors of severe COVID-19 to complete the genetic profiles of identified risk variants from public datasets and platforms. (C) The third step: analyzing molecular functions of risk factors of severe COVID-19 using gene ontology and Gene Card platforms. (D) The fourth step: constructing molecular networks and identifying functions and pathways between risk factors and other molecules from public datasets and platforms. (E) The fifth step: mapping constructed networks of severe COVID-19 with other diseases via shared risk variants using GWAS catalog, ClinGen., and other public resources


Distribution of the chromosomal location of risk variants

Fig. 2
figure 2

Chromosomal loci and functional consequences of the 109 genetic risk variants. Autosomal loci of the 60 risk genes associated with severe COVID-19: each dot represents a risk variant, and the dots in the same horizontal line represent the same risk variant but in different genes. The colors of the dots represent risk genes

The distribution of the 109 risk variants after filtration on the human chromosomes is presented in Fig. 2. The ideogram shows chromosomes 1-9, 11, 12, 15, 17, and 19-X contain 109 risk variants with different types of variant function such as intergenic, intronic, and transcript variants that are distributed in 60 genes in the host genome. We found two gene clusters that have an impact on severe outcomes of COVID-19. The first gene cluster contains C4BPA, LOC107985251, CFH, and CD55 located on chromosome 1q31.3-1q.32.2. The loci hold cluster of risk genes are related to immune responses and chemokine and cytokine activity. This cluster contains the regulatory variants rs61821041 and rs61821114, which contribute to downregulation of CD55. Furthermore, rs45574833 is associated with atypical hemolytic uremic syndrome, a condition in which thrombi develop in tiny blood arteries in the kidneys. The regulator of complement activation (RCA) contains many tandemly clustered genes with homologous immune system activities, which are downregulated during COVID-19 infection [15]. In addition, the second cluster of genes: SCN5A, LZTFL1, SLC6A20, FYCO1, CCR9, CXCR6, and XCR1 located on 3p21.31 and 3p22.2 are related to innate immune system activities. For instance, SCN5A in human macrophages functions as a pathogen sensor and modulates antiviral responses and defense [16], the cluster is carried by 50% of South Asian and 16% of European populations, and was previously associated with severe COVID-19 and immune dysfunction inherited from Neanderthals [17, 18]. Therefore, our investigation further emphasizes that this cluster is associated with severe COVID-19.

Moreover, the ideogram of the chromosomal locations of identified risk variants associated with severe COVID-19 illustrates that three haplotypes are located on chromosomes 9, 11, and X. Our investigation, we found that these haplotypes influence the immune response and metabolic system. Loci 9q34.3 holds a haplotype that contains six risk variants located in the ABO gene. This haplotype interacts with FUT1-6 and FUT and negatively influences the biosynthesis of the components of the blood group systems during COVID-19 infection. Additionally, a second IFITM3 haplotype on chromosome 11 was previously associated with higher severity of HIV, Dengue, Ebola, and influenza infections. IFITM3 is an immune effector protein that is essential for both controlling cytokine production and limiting viral replication. The last haplotype shown on the ideogram is located in the TLR7 gene on chromosome X.

In addition, a group of risk variants located within the 21q22.11-21q22.3 loci influence down-regulation of the activity of the IFNAR2 and MX1 genes, which regulate the function of the interferon receptors and B cell-activating factor receptors that are involved in the immune response to viruses.

Furthermore, we found that a group of variants on chromosome 1 (rs1202980, rs60220284, and rs45574833) located in the C4BPA, LOC107985251, and TRIM46 genes are related to breast cancer. Additionally, the rs4341 and rs4343 risk variants located in the AEC gene on chromosome 17 are related to Alzheimer disease and hypertension. Moreover, rs429358 and rs481778 located in APOE and PLEKHA4 genes on chromosome 19 are related to Alzheimer disease. More explanation of disease-variant mapping is provided in the supplementary materials.

At the level of single polymorphisms related to severe COVID-19, most chromosomes hold risk variants related to the blood and immune system that impact the immune response and increase the severity of COVID-19 during infection. For example, the risk variants found on chromosome 2 are related to the innate immune system, which is the first step of defense and interaction. In addition, some risk variants impact the level of biomarkers in blood. For example, the risk variants on chromosome 1 increase blood pressure during infection, so the biomarker (D-dimer) level increases. Furthermore, some of these variants influence metabolic pathways that affect body systems. For instance, the risk variant on chromosome 3 negatively impacts glucose metabolism pathways. Also, the length of the chromosome, gene size, and overlapping might affect variant distribution by chromosome.

Statistical analysis on the curated dataset of genetic risk variants of severe COVID-19 outcomes

In an initial analysis, our curated dataset of risk genetic variants of severe COVID-19 outcomes was analyzed with the list of risk variants with their reported effects. The data has 109 risk variants with 16 rare variants with minor allele frequencies (MAF) less than 0.01 and 93 common variants, as reported in the reviewed papers. Most of the reviewed papers reported MAF, P value, and other features. However, not all reviewed articles provided the odd ratio (OR) values of variants reported risk variants. Around 50% of the genetic effects of risk variants are not reported in the reviewed articles and need to be calculated. Thus, we re-estimated the additive effect of risk variants by calculating a new OR value for all risk variants using the reported OR and the MAF of each risk variant. Moreover, the association between the MAFs and genetic effects of risk variants of severe COVID-19 is visualized and explained in Additional file 1.

Fig. 3
figure 3

The additive effects of common risk variants on severe COVID-19 outcome per genes. The scatter plot shows the additive effects of common risk variants on severe COVID-19 outcomes per gene. Each point corresponds to the additive effect of the risk gene that has been calculated based on cumulative values of reported ORs and MAF. Each gene hosts at least one reported risk variant

The associations between risk variants and their effects in developing severe outcomes of COVID-19 per gene were established using an additive genetic model estimating the OR of severe COVID-19. The cumulative values of reported OR and MAF of the accumulated risk alleles in a particular gene were calculated and used as a combined effect to estimate the additive effects per gene. Figure 3 shows a scatter plot of the additive effects of risk variants on severe COVID-19 outcomes per gene. Moreover, the statistical method for estimating the genetic effects of severe COVID-19 is explained in Additional file 1.

Enrichment analysis of genetic risk factors for severe COVID-19

Our curated dataset represents variant profiles that provide descriptive information on risk variants. In our analysis, we named each variant associated with severe COVID-19 in the list as a risk variant and its host gene, as a risk gene causally associated with increased mortality in COVID-19. Furthermore, we named the proteins encoded by risk genes risk proteins. The list of genetic risk variants associated with severe COVID-19 is displayed in Additional file 2: Table S1. Below, we present the enrichment analysis of the genetic risk factors associated with the severity of COVID-19 at different levels: variants, genes, and proteins.

Variant level: COVID-19 risk variants and disease association

We found 46 risk variants in coding regions and 63 located in non-coding regions. In non-coding regions, three variants are located in the intergenic region between genes. Fifty-two of the 63 non-coding region variants are intronic variants, and the remainder are transcript variants occurring within intron regions [19, 20]. The distribution of the functional consequences of 109 filtered risk variants in human DNA is shown in Fig. 4.

Fig. 4
figure 4

Distribution of functional consequences of the 109 genetic risk variants in the human genome. The height of each bar represents the total number of risk variants in genetic regions, and the light gray color illustrates the number of risk variants associated with other diseases

Table 1 List of the risk variants for severe COVID-19 and their associated diseases and biomarkers

Table 1 displays the variant-disease associations for 41 risk variants. Seven risk variants are related to metabolic disorders such as diabetes mellitus, hyperglycemia, and blood protein levels. Three of the intron variants have an impact on ABO protein synthesis, which influences the red cell count and is associated with metabolic diseases such as diabetes and venous thromboembolism (VTE) [21, 22]. The rs12683493 risk variant changes glycosylation and causes von Willebrand disease [22, 23].

Moreover, four risk variants are linked with cardiovascular disorders such as cardiac arrhythmia, long QT syndrome, and triple vessel diseases. Three risk variants are associated with immune dysfunction; for example, white blood cell disorders, complete blood count disorders, and rheumatoid arthritis. Three risk variants are mapped to gastrointestinal diseases, such as Crohn’s disease and inflammatory bowel disease. Four risk variants are associated with cancer such as prostate cancer and Kaposi’s sarcoma. Another five risk variants are related to severe symptoms in infectious diseases namely, tuberculosis and severe influenza. One variant is related to respiratory disorders and mapped to severe asthma.

Gene level: COVID-19 risk genes and disease association

Fig. 5
figure 5

Risk gene set enrichment analysis of the 60 risk genes related to severe COVID-19. A The top ten ranked gene expression scores in human normal tissues and systems: the vertical axes represent the top ten systems based on ranked scores. Each bar represents a system, and the slots inside the bar represent the percentage of the risk genes expressed in various human tissues or cells based on the ranked gene expression scores. The horizontal axes represent the ranked score from 0 to 5. B The top ten human compartments and tissues with the highest numbers of enriched risk genes based on gene ontology analysis

The results of gene set analysis of the COVID-19 risk genes are shown in Fig. 5. The gene expression of the top ten ranked systems in normal tissues and cells derived from our enrichment analysis of the risk gene set is shown in the bar chart in Fig. 5A. Most of these risk genes are highly expressed in blood cells in the hematopoietic system, with a ranked score of 3.5 related to the immune response and viral interaction. The second system is the musculoskeletal system with a ranked score of 3, with some of the risk variants involved in blood circulation. The other three related systems are the renal, reproductive, and neuro systems and are approximately similar with ranked scores of 2.5. The risk genes for the respiratory system and the lungs only have a ranked score of 1.5.

Screening the list of risk genes can help to comprehend the organs and systems affected in severe COVID-19. However, the presence of gene expression does not necessarily mean that the gene is connected functionally to a network related to its expressed system or tissue. For example, three genes out of 60 risk genes are RNA-encoding genes affiliated with the ncRNA class LOC105378861, LOC107986083, and LOC107985251. The phenotype of LOC105378861 is related to the levels of tissue factor activity in blood and increased D-dimer levels in patients with COVID-19 [24]. However, no information is available about LOC107986083 and LOC107985251.

Figure 5B illustrates the number of risk genes that are enriched in human compartments and tissues. Mainly, most risk genes are enriched in the immune system and are located on the cell surface and related to the receptor of type I interferon or viral assembly compartments.

Fig. 6
figure 6

Gene set and disease analysis of the 60 risk genes related to the severity of COVID-19 outcomes. The top diseases were mapped to the risk genes based on the OMIM and Alliance-DISEASES databases. The circle size represents the number of genes associated with the disease, and the range of colors (high-significant level: red, low-significant level: green) represents the FDR scores for the disease associations

The results of the risk gene analysis with regard to disease associations are illustrated in Fig. 6. The top most-enriched risk genes mapped to disease based on the Mendelian Inheritance in Man (OMIM) database [25]. Inherited Alzheimer disease was the top disease that mapped to the risk genes, followed by long QT syndrome, myocardial infections, metabolic diseases, and lung dysfunction.

Protein level: COVID-19 risk proteins and molecular functional analysis

We found 56 proteins involved in the development of severe COVID-19. We applied GO enrichment analysis of the risk proteins derived from our dataset after eliminating three genes that did not encode proteins. This analysis showed 24 proteins are involved in the immune system.

Fig. 7
figure 7

GO enrichment analyses of the 56 risk proteins related to severe COVID-19 outcomes. A The top ten significant biological processes and hierarchical correlation clustering trees of the biological processes enriched with the risk proteins. B The top ten significant molecular functions and hierarchical correlation clustering trees of the molecular functions enriched with the risk proteins. C The top ten significant cellular components and hierarchical correlation clustering trees of the cellular components enriched with the risk proteins

Figure 7A shows the risk proteins are mainly enriched in the following biological processes: negative regulation of complement activation, response to interferon-beta, type I interferon signaling pathways, and cellular response to type I interferon. In terms of molecular function, Fig. 7B illustrates the risk proteins are highly enriched in peptidyl-dipeptidase activity, and chemokine binding with high false discovery rate (FDR) scores; these activities are involved in the immune system response. In addition, the results of the enrichment analysis of risk proteins in terms of cellular components are shown in Fig. 7C. The risk proteins are enriched in blood microparticles. Furthermore, the clustering trees summarize the correlation between the significant molecular pathways and functions. The pathways with many shared risk proteins cluster together in one branch. More significant P values are indicated by larger circles.

Table 2 Main host biological processes associated with the 56 risk proteins that contribute to the development of severe COVID-19

Table 2 displays the groups of risk proteins involved in biological process pathways related to immune and metabolic activities that contribute to the development of severe COVID-19 symptoms.

Summary of the enrichment analysis of risk factors for severe COVID-19

In summary, after applying a comprehensive enrichment analysis of our curated dataset of 109 risk variants associated with severe COVID-19 at different levels (variants, genes, and proteins), we mapped the genetic factors to related diseases to infer the relationships between severe COVID-19 and future complications. Based on the chromosomal distribution and risk variant-disease mapping, we identified three clusters of genes related to immune, hematopoietic, and metabolic dysfunction. Furthermore, three haplotypes contribute to hematopoietic and immune system complications. We found a haplotype of contiguous variants that contribute to regulatory functions or are associated with other diseases that cause severe COVID-19 symptoms. Six risk variants are located within contiguous loci in ABO involved in glycosylation metabolism. In addition, it is notable that the TMRSS2 and ACE2 receptors contain a massive number of contiguous variants that may increase the susceptibility of host cells to viral infection. Numerous groups of variants on chromosomes 21 and X influence the antigenic response of SARS-CoV-2 variants. Moreover, polymorphisms have an influence on host immune recognition and the susceptibility and intensity of the immune response to the SARS-CoV-2 virus [7, 24, 26,27,28,29]. Hence, we applied enrichment analysis on the set of host risk variants to evaluate gene differentiation and discover biomarkers for tissues and cells from the set. The results show that most genes are expressed in both the hematopoietic and immune systems. Furthermore, the molecular functions and biological processes of the list of proteins encoded by risk genes illustrate these proteins are involved in immune responses and metabolic activities.

Molecular network construction and integration

We eliminated four genes that did not encode proteins. Then, we constructed protein–protein interaction networks for 56 risk proteins that contain a total of 939 interacting proteins, including 48 risk proteins with 939 interactions, and seven orphan proteins that did not interact with any protein. We integrated all constructed networks for the risk proteins and obtained 24 connected PPI networks and seven orphan proteins. In addition, Table 3 displays the main systems and host tissues involved in the constructed networks and affected by the risk factors for severe COVID-19. More details of the PPI networks and related functions are displayed in Additional file 3.

Table 3 Summary of the 24 molecular networks constructed using the genetic risk factors for severe COVID-19

Based on the functional analysis above, most networks are related to the blood and immune systems. However, there is no apparent interaction that connects the proteins in these networks, which points to a missing interaction or link. Thus, to highlight such missing links, we adopted another approach to correlate the constructed networks with common diseases that are related to the risk variants hosted in these networks.

Fig. 8
figure 8

Molecular networks of the 56 risk proteins mapped with the 109 risk variant-disease associations. The red nodes represent COVID-19 risk proteins. The gray nodes represent human proteins that interact with risk proteins. A Twenty-four connected networks. Each connected network has at least one interaction with another protein. Networks 1 to 3 contain more than one risk protein, and Networks 4 to 22 are isolated networks has one risk protein. B Seven orphan risk proteins did not have any protein interaction with other human proteins. The dashed lines link risk variants with linked networks. The different colors of squares illustrate the disease mapping based on the risk variant-disease mapping and the similarity of the molecular functions between the constructed networks

Figure 8 illustrates the 24 PPI constructed based on the results of the risk variant-disease mapping in order to indirectly infer the common molecular functions between the constructed PPI networks. The variant-disease mapping is listed in Additional file 2: Table S2.

According to our investigation of the 24 constructed PPI networks, we infer that the molecular functions of the risk protein interactions are involved in three main host systems: the immune, metabolic, and cardiovascular systems.

In terms of risk variant mapping, Networks 2, 7, and 21 have a common risk variant rs13168774 located on chromosome 5 and associated with respiratory inflammation and metabolic disorders. Moreover, Networks 18, 19, 23, and 24 have a common risk variant rs11385942. These networks have common molecular functions related to immune responses. With reference to the similarity of biological processes and molecular functions between networks, Networks 1, 2, 9, 10, 15, 16, and 19-24 are related to the immune and hematopoietic systems. Networks 3, 7, 13, 14, and 17 are related to the metabolic system. Networks 4, 5, 17, and 18 are related to the cardiovascular system. Networks 3 and 17 are related to the urinary system. Network 11 is related to the nervous system. Network 12 is related to the endomembrane system.

Based on the evidence gleaned from the molecular function analysis of the constructed networks and risk variant-disease mapping, we inferred the hidden PPIs between unconnected networks. The molecular function analysis revealed several hidden pathways. For example, we found unconnected Networks 1-3, 6-10, and 12-24 have similar molecular functions and biological processes related to the immune system. In addition, we inferred unconnected Networks 1-5, 13, 14, and 17-16 share common pathways related to the cardiovascular system and diseases. Moreover, Networks 1, 3, and 11 have common functions and diseases related to the neuron system. Furthermore, based on the disease mapping similarity, we derived that unconnected Networks 1-3, 15-17, and 21 have pathways related to the renal system. Thus, we infer hidden interactions could connect these networks.

Molecular pathways of constructed networks

Out of the 24 constructed networks, we analyzed the pathways in Network 1, the largest network that has the highest number of risk proteins (n = 16) and pathways related to the host immune system and contains 511 interactions. Based on the top ten functional pathways of Network 1, we found that eight of the ten molecular pathways in Network 1 are related to cytokine signaling and responses in the immune system and SARS-CoV-2 responses in the innate immune system. The remaining three pathways are related to Influenza A, measles, and Hepatitis C infections. Figure 9 shows the molecular pathways for the risk factors from Network 1; this network is related to the immune system, which confirms that the majority of pathways in Network 1 are related to immune responses. Moreover, the functional pathways in Network 2 contain five risk proteins and 75 interactions that are mainly related to the innate immune system. We found that Network 2 has eight significant pathways with P values higher than 2.6e−13 related to the complement and coagulation cascade pathways. The complement system is a central component of the innate immune system.

Fig. 9
figure 9

Molecular pathways between the 16 genetic risk proteins of severe COVID-19 in Network 1. The top ten significant molecular pathways between the genetic risk proteins and other host proteins are mainly connected to the host immune system. Network 1 is the largest network and contains the highest number of risk proteins compared to other networks

However, some constructed networks are not related to the immune system. For instance, Network 3 contains three risk proteins and 27 interactions and is related to the metabolic and renin-angiotensin systems, which play an essential role in the regulatory functions and processes of renal, cardiac, and vascular metabolism and physiology. We found that Network 3 has five significant pathways with P values < 5e−08 related to peptide hormone metabolism and protein digestion and absorption, which are involved in metabolic processes and systems. In addition, Network 4 contains three risk proteins and 27 interactions related to the cardiovascular system. We found that all top ten pathways in Network 4 are involved NOTCH signaling and the intracellular domain of NOTCH regulates transcription of genes related to cardiac development. Figure 10 shows the molecular pathways of the remaining constructed Networks 2-24. More details of the molecular pathways between the proteins in these networks and pathway analysis of the remaining constructed networks are demonstrated in Additional files 3 and 4.

Fig. 10
figure 10

Overview of the molecular pathways of the risk proteins in the constructed networks related to the host metabolic, cardiovascular, and other systems. The figure demonstrates the significant pathways related to severe COVID-19 outcomes. More details of the remaining constructed networks are provided in Additional file 4

Host genetic risk factors and the SARS-CoV-2 pathway

After applying molecular function enrichment analysis and disease mapping to risk factors related to severe COVID-19, we found that the majority of pathways are related to the immune system, which suggests that the risk factors associated with COVID-19 are mostly present in proteins involved in the immune system. A minority of networks that have three or more risk proteins, such as Networks 3 and 4, are related to the metabolic and cardiovascular systems. This evidence indicates that the genetic risk factors associated with severe COVID-19 are involved in different host systems that cause multi-organ dysfunctions.

Fig. 11
figure 11

Host-SARS-CoV-2 pathogen interaction pathway in COVID-19. The location of the 56 genetic risk proteins in the main molecular pathway involved in the host response to SARS-CoV-2 derived from KEEG database. The proteins involved in this pathway are components of different constructed molecular networks, such as Network 1, 3, and 20

Figure 11 shows the location of the risk proteins in the main host-virus pathway, which contains ten risk proteins from different constructed molecular networks. For instance, IL-6, TLR7, OAS, and IFNAR from Network 1 and TMPRSS2 from Network 20 are mainly related to immune system and cytokine and interferon signaling. In addition, ACE and ACE2 from Network 3 are related to the metabolic system and processes.


The limited understanding of biological mechanisms and the impact of host genetic risk factors on severe COVID-19 has led researchers to identify genetic risk variants and analyze their influence on the development of severe symptoms. Our study is analogous to the molecular network analysis studies of protein interactions derived from risk variants identified from GWAS and statistical summaries conducted among different populations [3, 30, 31].

Moreover, the use of molecular networks has facilitated progress in many areas of biomedical science, such as understanding and linking the functional molecular interactions between proteins with potential targets that could lead to drug development. This straightforward and effective concept allows us to extract the core relationships between host genes and proteins and it also helps to identify and predict interactions between drugs, study the comorbidity of disease, and discover essential associations. Thus, we constructed the PPI networks for risk proteins associated with severe COVID-19 and inferred the hidden interaction based on disease mapping and GO functional analysis.

Our results are in general agreement with previous association studies of host genetic variants related to severe symptoms during COVID-19 infection. We showed that the identified genetic risk variants and their molecular functions are involved in several activities, mainly related to the immune system, along with notable relationships to the metabolic and hematologic systems. These systems react to viruses and prevent the host from developing severe symptoms during infection, thus the risk variants could lead to multi-organ dysfunction since these systems are involved in the activities of numerous genes expressed in various organs in the human body. Following SARS-CoV-2 infection, the host body activates a complex regulatory system of innate responses to defend against the virus. We found that 14 out of the 22 PPI networks influence cytokine and chemokine signaling; for example, risk factors in the pro-inflammatory cytokine genes IL-6 and IL-10 contribute to the process of pathological pain activities in the immune system. In addition, some risk proteins such as APOE, CD55, C4BPA, and CFH are involved in severe changes in the interferon alpha/beta signaling pathways, which play a vital function in the host immune response to viruses and help to prevent SARS-CoV-2 infection.

Although the majority of biological pathways and activities between the constructed networks are related to the immune system, we observed noticeable biological pathways that negatively contribute to the metabolic and hematopoietic systems, and development of multi-organ dysfunction such as heart attack, liver and kidney injury, and lung inflammation. We found that 6 out of 22 constructed networks are centrally related to the renin-angiotensin-aldosterone system (RAAS) that regulates the production of the hormone angiotensin II. This hormone binds to its receptors in the human host tissue and has various impacts on several organ activities, such as stimulating vasal construction in the arterioles and promoting sodium reabsorption in the kidneys [32].

Moreover, the RAAS contributes to the central nervous system, which suggests that angiotensin II could have effects on the brain. For example, angiotensin II stimulates thirst by acting on the hypothalamus. Furthermore, angiotensin II reduces the baroreceptor response to increased blood pressure, so that this response would not counteract the effect of RAAS. Overall, the effects of the RAAS on metabolic and blood systems lead to increased blood volume and pressure. Also, some of the risk proteins for severe COVID-19 are present in components of the RAAS that are used as clinical biomarkers related to the hematopoietic system. In addition, some biomarkers of RAAS such as ACE are affected by risk variants located in ACE2 [33, 34]. These risk variants downregulate ACE2 expression, which affects the ACE inhibitor and bradykinin pathways and increases activities of multiple pathophysiological pathways that contribute to cardiovascular disease. Furthermore, hypertension is caused by inappropriate activation of the RAAS. Hence, this system is often a target for antihypertensive drugs and the levels of some RAAS proteins are used as medical blood markers during diagnosis of COVID-19 [35,36,37]. Thus, taking genetic risk factors into account during the diagnosis of patients with COVID-19 could help to prevent organ complications by enabling future complications to be predicted at an early stage based on genetic analysis of risk factors. This suggestion is based on our deep systematic analysis workflow of all genetic risk factors associated with severe COVID-19 from scientific articles and medical reports published during the last two years. Overall, our enrichment analysis and construct PPI networks of genetic risk factors. associated with severe COVID-19 support the value of personalized medicine approaches to predict future complications and organ dysfunction during or after infection among patients with COVID-19.


In conclusion, our study highlights the potential of studying the molecular functions and interactions of genetic risk factors to identify significant biological mechanisms and pathogenicity pathways that are involved in the development of severe COVID-19 outcomes. Moreover, the study emphasizes the importance of inferring the hidden interactions between networks based on the disease and functional similarity between the networks. We found most of the pathways discovered, and hence the associated risk factors, are mainly related to the immune system with a notable number of pathways related to the metabolic and cardiovascular systems. This evidence reveals the genetic risk factors associated with COVID-19 are involved in several pathways that cause multi-organ dysfunction in different systems. This work underscores the importance of analyzing the molecular interactions and pathways between the SARS-CoV-2 virus and the host to understand the heterogeneous susceptibility of the immune response. This study proposes new insights into pathogenicity analysis of infections by including risk genetic information as essential factors to support individualised clinical treatment plans and predict future complications during and after infection.

Materials and methods

We designed a computational workflow to apply a deep systematic analysis of identified variants related to severe COVID-19 outcomes that were reported in 35 published works from December 2019 to 2021. These 35 studies were identified from the PubMed search engine using keywords queries such as “Severe-COVID-19”. The 125 variants were retrieved from these genetic studies using text mining algorithms and we validated the final list of risk variants manually. The list of risk variants passed through four phases: deep enrichment analysis of the molecular functions and pathways the risk factors are involved in, and construction of molecular networks to understand the pathogenicity and mechanisms that lead to severe COVID-19 outcomes.

Characteristic of retrieved dataset of COVID-19 risk factors

From GWAS studies of all loci in the host human genome with a sample size more than or equal 500 [3, 7, 24, 26,27,28, 30, 38, 39], we retrieved and listed the genetic variants associated with severe COVID-19 in any population with P values \(< 5 \times 10^{-5}\) or risk variants mentioned as significant in review studies. Most of the retrieved articles were case-control studies of the genome sequences patients with COVID-19. The papers included patients of Chinese, European, Middle-Eastern, and other ethnicities, with different symptoms and comorbidities, with an approximate average age of 63 for males and 58 for females. The youngest patient was only two-years-old, and the oldest was an 85-year-old man. Most papers classified the patients into six levels of severity: asymptomatic, mild, moderate, severe, critical, and deceased. Two articles conducted GWAS with controls who had mild/moderate symptoms to avoid type II error. Most variants identified in GWAS or statistical analysis had 95% significance and above. However, not all retrieved papers mentioned the P values for the identified variants. Since the final number of filtered variants was relatively small and not all papers indicated P values, we neglected the condition of significance. In our analysis, we called each variant causally associated with severe COVID-19 a risk variant and its host gene, a risk gene. Furthermore, we called the proteins encoded by risk genes risk proteins.

Computational workflow

Fig. 12
figure 12

Computational network analysis workflow. The four-stage implemented workflow: (A) data curation and annotation, (B) functional enrichment analysis of risk factors, (C) molecular network construction and integration of risk factors, and (D) molecular network analysis and mapping

A general overview of the phases involved in our computational network analysis workflow is shown in Fig. 12. Our workflow contains four phases: data curation and annotation, functional enrichment analysis of risk factors, molecular network construction and integration of risk factors, and molecular network analysis and mapping based on related diseases and the similarity of ontologies of the risk factor pathways and their molecular functions.

Phase 1: data curation and annotation

In phase one, we retrieved 125 risk genetic variants associated with severe COVID-19 from GWAS studies and statistical genetic studies based on GWAS reports. After quality control and removing duplicates, 109 risk variants were annotated using known databases. The dbSNP database [40] was used to identify the chromosomal location of the risk variants based on the CRCh 38 human reference genome for consistency purposes and Ensembl [41] was used to retrieve genomic information and the types of the variants.

In addition, the profiles of the 109 genetic risk variants were mapped to 60 genes using the GeneCards platform [42, 43]. Then, the risk variants were mapped to related diseases using ClinVar [44], Online OMIM [25], and MalaCards databases [45] to complete the profile of each variant to give intensive information.

The curated dataset of genetic risk factors for severe COVID-19 contains 109 risk variants, 60 risk genes, and 56 proteins, and other features that provide a fingerprint of each risk factor related to the severity of COVID-19. Eleven annotations were used to describe the risk variants, including the main features (rs ID, chromosomal location, functional consequence, host gene, P value, related diseases, and references). The curated dataset is provided in Additional file 2: Table S1.

Phase 2: functional enrichment analysis of risk factors

In phase two, enrichment functional analyses were applied on our curated dataset using Gene Ontology (GO) [46] on the risk variants, genes, and proteins to obtain enriched information describing the molecular functions and processes of the genetic risk genetic factors related to severe COVID-19.

In terms of risk variant analyses, we used the Ensembl [41] and VarElect [47] platforms to identify the structure and functional consequences of the risk variants to obtain the functional distribution and characteristics of the risk variants.

At the risk gene level, we applied expression-based analysis using GeneAnalytics [48], which relies on LifeMap Discovery [49] to identify gene associations with tissues and cells. The matching scores for gene expression in different normal tissues and cells were calculated with a search for maximum similarity between expression vectors [48]. The ranked scores for enriched risk genes were calculated using LifeMap discovery by integrating gene annotations from scientific publications, as well as bioinformatic data. A matching score was assigned to each risk gene based on their annotations in the data. The overall score depends on the number of matches for each risk gene. Moreover, we assessed the matching quality by categorizing the results as high-, medium-, and low-quality matches; only high-quality matches were considered in the analysis. The distribution of the matching qualities across the list of results is also presented to provide an easy assessment. Then, we applied functional enrichment analysis to the list of risk genes related to severe COVID-19 outcomes using GO terms and GeneAnalytics [40, 48]. We used ShinyGo [50] to analyze the ontologies of the compartments and tissues of the risk genes and associated diseases based on a 0.05 FDR cutoff and minimum and maximum pathway size of 2-2000.

In terms of risk protein analyses, we used STRING 11.5 [51] and ShinyGo [50] to identify molecular functions and biological pathways associated with the 56 risk proteins based on a threshold of FER equal 0.05.

Phase 3: molecular network construction and integration

In phase three, we constructed and analyzed the PPI networks of the 56 risk proteins using the STRING 11.5 database [51, 52]. All constructed networks are full STRING networks, which means the edges between proteins indicate both functional and physical associations. Having a functional association means that the proteins have the same function, while having physical associations indicate common molecular components that link the proteins. We considered all active sources of protein interactions that are reported in the literature, validated experimentally, and retrieved from public databases such as GeneCards for gene interpretations, Reactome [53], and Kyoto Encyclopedia of Genes and Genomes (KEGG) [54, 55] for gene pathways.

Furthermore, all retrieved interactions satisfy the criterion of having an interaction score of at least 0.9 to guarantee that the associated interaction is sufficiently significant to enhance further analysis. We constructed the top ten interactions for each risk protein based on the highest significant P values among all PPI candidates to limit the network outputs and avoid unclean networks with unnecessary interfering links for more effective network visualization.

Then we integrated the PPI networks constructed for the 56 risk proteins with the 109 risk variants. Visualization of the integrated data using Cytoscape [56] generated 24 PPI networks and seven orphan proteins. Then we mapped the unconnected networks based on their common risk variants or the similarity of their molecular functions using GO, STRING database [51, 52], and GeneCards.

Phase 4: molecular network pathway analysis and disease mapping

Finally in phase four, we analyzed the molecular network pathways to identify pathogenicity and pathways that related to severe outcomes of COVID-19 using Reactome [53] and the Kyoto Encyclopedia of Genes and Genomes (KEGG) [54, 55]. In addition, we mapped the correlation between disease-variant-network to find similarity and shared biological pathways between related disease and the molecular functions of the constructed networks using the ClinVar [44], PathCards [57], and MalaCards platforms. Then, we linked the outputs to infer the hidden connections between the constructed networks based on disease mapping, network molecular functions, gene ontology analysis, and pathway similarity.

Availability of data and materials

The curated dataset of genetic risk factors of severe COVID-19 is provided in the Additional file 1.



Severe Acute Respiratory Syndrome Coronavirus 2


Coronavirus 2019 disease


World Health Organization


Centers of Disease and Control Preventions


Exome-Wide Association Study


Genome-Wide Association Study


Mendelian randomization


Bayesian colocalization


Phenome-Wide Association Study


Expression quantitative trait loci


Single nucleotide polymorphisms


Transcriptome-Wide Association Study


Whole genome sequencing


Gene expression quantitative trait loci


Protein quantitative trait loci


\(3^\prime \) Untranslated region


False discovery rate


Protein–protein interaction


Transcription factor


Gene ontology


Kyoto encyclopedia of genes and genomes


Major histocompatibility complex


Basic helix-loop-helix


Systemic lupus erythematosus


Aicardi–Goutières syndrome


Bare lymphocyte syndrome, type I


Development of atherosclerotic cardiovascular disease


Coronary artery disease


Venous thromboembolism


Thyroid-stimulating hormone


Mean corpuscular hemoglobin concentration


Red blood cell


Hemolytic uremic syndrome


Long QT syndrome


Intracerebral hemorrhage


Parathyroid hormone


Renal tubular dysgenesis


Regulator of complement activation


Renin–angiotensin–aldosterone system


  1. Wang H, Li X, Li T, Zhang S, Wang L, Wu X, Liu J. The genetic sequence, origin, and diagnosis of SARS-CoV-2. Eur J Clin Microbiol Infect Dis. 2020;39(9):1629–35.

    Article  CAS  PubMed  PubMed Central  Google Scholar 

  2. Shi Y, Wang G, Cai X-P, Deng J-W, Zheng L, Zhu H-H, Zheng M, Yang B, Chen Z. An overview of COVID-19. J Zhejiang Univ Sci B. 2020;21(5):343–60.

    Article  CAS  PubMed  PubMed Central  Google Scholar 

  3. Hu J, Li C, Wang S, Li T, Zhang H. Genetic variants are identified to increase risk of COVID-19 related mortality from UK biobank data. Hum Genomics. 2021;15(1):1–10.

    Article  Google Scholar 

  4. Ramos-Lopez O, Daimiel L, Ramírez de Molina A, Martínez-Urbistondo D, Vargas JA, Martínez JA. Exploring host genetic polymorphisms involved in SARS-CoV infection outcomes: Implications for personalized medicine in COVID-19. Int J Genomics 2020;2020.

  5. Butler-Laporte G, Nakanishi T, Mooser V, Morrison DR, Abdullah T, Adeleye O, Mamlouk N, Kimchi N, Afrasiabi Z, Rezk N, et al. Vitamin D and COVID-19 susceptibility and severity in the COVID-19 host genetics initiative: a mendelian randomization study. PLoS Med. 2021;18(6):1003605.

    Article  Google Scholar 

  6. Hernández Cordero AI, Li X, Milne S, Yang CX, Bossé Y, Joubert P, Timens W, van den Berge M, Nickle D, Hao K, et al. Multi-omics highlights abo plasma protein as a causal risk factor for COVID-19. Hum Genet. 2021;140(6):969–79.

    Article  PubMed  PubMed Central  Google Scholar 

  7. Dai Y, Wang J, Jeong H-H, Chen W, Jia P, Zhao Z. Association of CXCR6 with COVID-19 severity: delineating the host genetic factors in transcriptomic regulation. Hum Genet. 2021;140(9):1313–28.

    Article  CAS  PubMed  PubMed Central  Google Scholar 

  8. Consortium GTE, et al. The GTEx consortium atlas of genetic regulatory effects across human tissues the genotype tissue expression consortium. Science. 2019;369(6509):1318–30.

  9. Malkova A, Kudlay D, Kudryavtsev I, Starshinova A, Yablonskiy P, Shoenfeld Y. Immunogenetic predictors of severe COVID-19. Vaccines. 2021;9(3):211.

    Article  CAS  PubMed  PubMed Central  Google Scholar 

  10. Bastard P, Rosen LB, Zhang Q, Michailidis E, Hoffmann H-H, Zhang Y, Dorgham K, Philippot Q, Rosain J, Béziat V, et al. Autoantibodies against type I IFNs in patients with life-threatening COVID-19. Science. 2020;370(6515):4585.

    Article  Google Scholar 

  11. Zhang Q, Bastard P, Liu Z, Le Pen J, Moncada-Velez M, Chen J, Ogishi M, Sabli IK, Hodeib S, Korol C, et al. Inborn errors of type I IFN immunity in patients with life-threatening COVID-19. Science. 2020;370(6515):4570.

    Article  Google Scholar 

  12. Zhang Q, Bastard P, Cobat A, Casanova J-L. Human genetic and immunological determinants of critical COVID-19 pneumonia. Nature. 2022;603(7902):587–98.

    Article  CAS  PubMed  PubMed Central  Google Scholar 

  13. Beck DB, Aksentijevich I. Susceptibility to severe COVID-19. Science. 2020;370(6515):404–5.

    Article  PubMed  Google Scholar 

  14. Moon CY, Schilder BM, Raj T, Huang K-L. Phenome-wide and expression quantitative trait locus associations of coronavirus disease 2019 genetic risk loci. Iscience. 2021;24(6): 102550.

    Article  CAS  PubMed  PubMed Central  Google Scholar 

  15. Cheng J, Clayton JS, Acemel RD, Zheng Y, Taylor RL, Keleş S, Franke M, Boackle SA, Harley JB, Quail E, et al. Regulatory architecture of the RCA gene cluster captures an intragenic tad boundary, CTCF-mediated chromatin looping and a long-range intergenic enhancer 2020.

  16. Jones A, Kainz D, Khan F, Lee C, Carrithers MD. Human macrophage SCN5A activates an innate immune signaling pathway for antiviral host defense. J Biol Chem. 2014;289(51):35326–40.

    Article  CAS  PubMed  PubMed Central  Google Scholar 

  17. Lee J-W, Lee I-H, Sato T, Kong SW, Iimura T. Genetic variation analyses indicate conserved SARS-CoV-2-host interaction and varied genetic adaptation in immune response factors in modern human evolution. Dev Growth Differ. 2021;63(3):219–27.

    Article  CAS  PubMed  PubMed Central  Google Scholar 

  18. Zeberg H, Pääbo S. The major genetic risk factor for severe COVID-19 is inherited from Neanderthals. Nature. 2020;587(7835):610–2.

    Article  CAS  PubMed  Google Scholar 

  19. Griffiths JF, Griffiths AJ, Wessler SR, Lewontin RC, Gelbart WM, Suzuki DT, Miller JH, et al. An introduction to genetic analysis. Macmillan; 2005.

  20. Frézal J, Bonaïti-Pellié C. Introduction to genetic analysis. In: Principles and prenatal growth, pp. 229–247. Springer; 1978.

  21. Reily C, Stewart TJ, Renfrow MB, Novak J. Glycosylation in health and disease. Nat Rev Nephrol. 2019;15(6):346–66.

    Article  PubMed  PubMed Central  Google Scholar 

  22. Deng H, Yan X, Yuan L. Human genetic basis of coronavirus disease 2019. Signal Transduct Target Ther. 2021;6(1):1–14.

    CAS  Google Scholar 

  23. Whiffin N, Karczewski KJ, Zhang X, Chothani S, Smith MJ, Evans DG, Roberts AM, Quaife NM, Schafer S, Rackham O, et al. Characterising the loss-of-function impact of 5’untranslated region variants in 15,708 individuals. Nat Commun. 2020;11(1):1–12.

    Article  Google Scholar 

  24. Ramlall V, Thangaraj PM, Meydan C, Foox J, Butler D, Kim J, May B, De Freitas JK, Glicksberg BS, Mason CE, et al. Immune complement and coagulation dysfunction in adverse outcomes of SARS-CoV-2 infection. Nat Med. 2020;26(10):1609–15.

    Article  CAS  PubMed  PubMed Central  Google Scholar 

  25. Hamosh A, Scott AF, Amberger JS, Bocchini CA, McKusick VA. Online mendelian inheritance in man (OMIM), a knowledgebase of human genes and genetic disorders. Nucleic Acids Res. 2005;33(1):514–7.

    Google Scholar 

  26. Li Y, Ke Y, Xia X, Wang Y, Cheng F, Liu X, Jin X, Li B, Xie C, Liu S, et al. Genome-wide association study of COVID-19 severity among the Chinese population. Cell Discov. 2021;7(1):1–16.

    Article  PubMed  PubMed Central  Google Scholar 

  27. Russo R, Andolfo I, Lasorsa VA, Cantalupo S, Marra R, Frisso G, Abete P, Cassese GM, Servillo G, Esposito G, et al. The TNFRSF13C H159Y variant is associated with severe COVID-19: a retrospective study of 500 patients from southern Italy. Genes. 2021;12(6):881.

    Article  CAS  PubMed  PubMed Central  Google Scholar 

  28. Wang F, Huang S, Gao R, Zhou Y, Lai C, Li Z, Xian W, Qian X, Li Z, Huang Y, et al. Initial whole-genome sequencing and analysis of the host genetic contribution to COVID-19 severity and susceptibility. Cell Discov. 2020;6(1):1–16.

    Article  Google Scholar 

  29. Group SC-G. Genomewide association study of severe COVID-19 with respiratory failure. N Engl J Med 2020;383(16):1522–34.

  30. Pathak GA, Singh K, Miller-Fleming TW, Wendt FR, Ehsan N, Hou K, Johnson R, Lu Z, Gopalan S, Yengo L, et al. Integrative genomic analyses identify susceptibility genes underlying COVID-19 hospitalization. Nat Commun. 2021;12(1):1–11.

    Article  Google Scholar 

  31. Iniguez M, Pérez-Matute P, Villoslada-Blanco P, Recio-Fernandez E, Ezquerro-Pérez D, Alba J, Ferreira-Laso ML, Oteo JA. Ace gene variants rise the risk of severe COVID-19 in patients with hypertension, dyslipidemia or diabetes: a Spanish pilot study. Front Endocrinol 2021;1041.

  32. Crowley SD, Coffman TM. Recent advances involving the renin–angiotensin system. Exp Cell Res. 2012;318(9):1049–56.

    Article  CAS  PubMed  PubMed Central  Google Scholar 

  33. Ingraham NE, Barakat AG, Reilkoff R, Bezdicek T, Schacker T, Chipman JG, Tignanelli CJ, Puskarich MA. Understanding the renin–angiotensin–aldosterone–SARS-CoV axis: a comprehensive review. Eur Respir J 2020;56(1).

  34. Lopes RD, Macedo AV, Silva PGDBE, Moll-Bernardes RJ, Dos Santos TM, Mazza L, Feldman A, Arruda GDS, Denílson C, Camiletti AS, et al. Effect of discontinuing vs continuing angiotensin-converting enzyme inhibitors and angiotensin II receptor blockers on days alive and out of the hospital in patients admitted with COVID-19: a randomized clinical trial. Jama. 2021;325(3):254–64.

    Article  CAS  PubMed  PubMed Central  Google Scholar 

  35. Sparks MA, Crowley SD, Gurley SB, Mirotsou M, Coffman TM. Classical renin-angiotensin system in kidney physiology. Compr Physiol. 2014;4(3):1201.

    Article  PubMed  PubMed Central  Google Scholar 

  36. Wang W, Zhao X, Wei W, Fan W, Gao K, He S, Zhuang X. Angiotensin-converting enzyme inhibitors (ACEI) or angiotensin receptor blockers (ARBs) may be safe for COVID-19 patients. BMC Infect Dis. 2021;21(1):1–8.

    Google Scholar 

  37. Chiong JR, Miller AB. Renin–angiotensin system antagonism and lipid-lowering therapy in cardiovascular risk management. J Renin Angiotensin Aldosterone Syst. 2002;3(2):96–102.

    Article  CAS  PubMed  Google Scholar 

  38. Latini A, Agolini E, Novelli A, Borgiani P, Giannini R, Gravina P, Smarrazzo A, Dauri M, Andreoni M, Rogliani P, et al. COVID-19 and genetic variants of protein involved in the SARS-CoV-2 entry into the host cells. Genes. 2020;11(9):1010.

    Article  CAS  PubMed  PubMed Central  Google Scholar 

  39. Novelli A, Biancolella M, Borgiani P, Cocciadiferro D, Colona VL, D’Apice MR, Rogliani P, Zaffina S, Leonardis F, Campana A, et al. Analysis of ACE2 genetic variants in 131 Italian SARS-CoV-2-positive patients. Hum Genomics. 2020;14(1):1–6.

    Article  Google Scholar 

  40. Sherry ST, Ward M-H, Kholodov M, Baker J, Phan L, Smigielski EM, Sirotkin K. dbSNP: the NCBI database of genetic variation. Nucleic Acids Res. 2001;29(1):308–11.

    Article  CAS  PubMed  PubMed Central  Google Scholar 

  41. Howe KL, Achuthan P, Allen J, Allen J, Alvarez-Jarreta J, Amode MR, Armean IM, Azov AG, Bennett R, Bhai J, et al. Ensembl 2021. Nucleic Acids Res. 2021;49(D1):884–91.

    Article  Google Scholar 

  42. Stelzer G, Rosen N, Plaschkes I, Zimmerman S, Twik M, Fishilevich S, Stein TI, Nudel R, Lieder I, Mazor Y, et al. The genecards suite: from gene data mining to disease genome sequence analyses. Curr Protoc Bioinform. 2016;54(1):1–30.

    Article  Google Scholar 

  43. Safran M, Rosen N, Twik M, BarShir R, Stein TI, Dahary D, Fishilevich S, Lancet D. Practical guide to life science databases. Singapore: The GeneCards Suite; Springer Nature Singapore 2021, pp. 27–56.

  44. Landrum MJ, Lee JM, Benson M, Brown G, Chao C, Chitipiralla S, Gu B, Hart J, Hoffman D, Hoover J, et al. Clinvar: public archive of interpretations of clinically relevant variants. Nucleic Acids Res. 2016;44(D1):862–8.

    Article  Google Scholar 

  45. Rappaport N, Twik M, Nativ N, Stelzer G, Bahir I, Stein TI, Safran M, Lancet D. Malacards: a comprehensive automatically-mined database of human diseases. Curr Protoc Bioinform. 2014;47(1):1–24.

    Article  Google Scholar 

  46. Consortium GO. The gene ontology (GO) database and informatics resource. Nucleic Acids Res. 2004;32(1):258–61.

  47. Stelzer G, Plaschkes I, Oz-Levi D, Alkelai A, Olender T, Zimmerman S, Twik M, Belinky F, Fishilevich S, Nudel R, et al. Varelect: the phenotype-based variation prioritizer of the genecards suite. BMC Genomics. 2016;17(2):195–206.

    Google Scholar 

  48. Ben-Ari Fuchs S, Lieder I, Stelzer G, Mazor Y, Buzhor E, Kaplan S, Bogoch Y, Plaschkes I, Shitrit A, Rappaport N, et al. Geneanalytics: an integrative gene set analysis tool for next generation sequencing, RNAseq and microarray data. Omics J Integr Biol. 2016;20(3):139–51.

    Article  CAS  Google Scholar 

  49. Edgar R, Mazor Y, Rinon A, Blumenthal J, Golan Y, Buzhor E, Livnat I, Ben-Ari S, Lieder I, Shitrit A, et al. Lifemap discovery’: the embryonic development, stem cells, and regenerative medicine research portal. PLoS ONE. 2013;8(7):66629.

    Article  Google Scholar 

  50. Ge SX, Jung D, Yao R. Shinygo: a graphical gene-set enrichment tool for animals and plants. Bioinformatics. 2020;36(8):2628–9.

    Article  CAS  PubMed  Google Scholar 

  51. Mering CV, Huynen M, Jaeggi D, Schmidt S, Bork P, Snel B. String: a database of predicted functional associations between proteins. Nucleic Acids Res. 2003;31(1):258–61.

    Article  Google Scholar 

  52. Szklarczyk D, Gable AL, Nastou KC, Lyon D, Kirsch R, Pyysalo S, Doncheva NT, Legeay M, Fang T, Bork P, et al. The string database in 2021: customizable protein–protein networks, and functional characterization of user-uploaded gene/measurement sets. Nucleic Acids Res. 2021;49(D1):605–12.

    Article  Google Scholar 

  53. Jassal B, Matthews L, Viteri G, Gong C, Lorente P, Fabregat A, Sidiropoulos K, Cook J, Gillespie M, Haw R, et al. The reactome pathway knowledgebase. Nucleic Acids Res. 2020;48(D1):498–503.

    Google Scholar 

  54. Kanehisa M, Goto S. Kegg: Kyoto encyclopedia of genes and genomes. Nucleic Acids Res. 2000;28(1):27–30.

    Article  CAS  PubMed  PubMed Central  Google Scholar 

  55. Ogata H, Goto S, Sato K, Fujibuchi W, Bono H, Kanehisa M. Kegg: Kyoto encyclopedia of genes and genomes. Nucleic Acids Res. 1999;27(1):29–34.

    Article  CAS  PubMed  PubMed Central  Google Scholar 

  56. Smoot ME, Ono K, Ruscheinski J, Wang P-L, Ideker T. Cytoscape 2.8: new features for data integration and network visualization. Bioinformatics. 2011;27(3):431–2.

    Article  CAS  PubMed  Google Scholar 

  57. Belinky F, Nativ N, Stelzer G, Zimmerman S, Iny Stein T, Safran M, Lancet D. Pathcards: multi-source consolidation of human biological pathways. Database 2015;2015.

Download references


We thank Mr. Mohammed Alarawi for the discussion on the molecular functions of protein networks, Dr. Marwa Abdelhakim for the discussion on the identified genetic haplotype and gene clusters of the list of risk variants, and Dr. Malak Alsaedi for the medical discussion on the impact of risk variants on developing severe outcomes related to metabolic and immune disease.


This work was supported by the King Abdullah University of Science and Technology (KAUST) Office of Sponsored Research (OSR) under Award No BAS/1/1059-01-01, FCC/1/1976-44-01, and FCC/1/1976-45-01.

Author information

Authors and Affiliations



S.A. designed the study experiments of analysing the molecular functions and pathways between risk factors and carried out the PPI molecular network analysis. S.A prepared the manuscript under the supervision of T.G., K.M. and X.G. helped to draft and review the manuscript. All authors read and approved the final manuscript.

Corresponding author

Correspondence to Takashi Gojobori.

Ethics declarations

Ethics approval and consent to participate

Not applicable.

Consent for publication

Not applicable.

Competing interests

The authors declare that they have no competing interests.

Additional information

Publisher's Note

Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations.

Supplementary Information

Additional file 1.

Details of estimating genetic effects of severe COVID-19 outcomes.

Additional file 2.

List of the 109 genetic risk variants and 24 network mapping for severe COVID-19.

Additional file 3.

Enrichment analysis of the molecular functions of the 24 constructed networks.

Additional file 4.

Details of the molecular pathways of the 24 constructed protein–protein interaction networks.

Rights and permissions

Open Access This article is licensed under a Creative Commons Attribution 4.0 International License, which permits use, sharing, adaptation, distribution and reproduction in any medium or format, as long as you give appropriate credit to the original author(s) and the source, provide a link to the Creative Commons licence, and indicate if changes were made. The images or other third party material in this article are included in the article's Creative Commons licence, unless indicated otherwise in a credit line to the material. If material is not included in the article's Creative Commons licence and your intended use is not permitted by statutory regulation or exceeds the permitted use, you will need to obtain permission directly from the copyright holder. To view a copy of this licence, visit The Creative Commons Public Domain Dedication waiver ( applies to the data made available in this article, unless otherwise stated in a credit line to the data.

Reprints and permissions

About this article

Check for updates. Verify currency and authenticity via CrossMark

Cite this article

Alsaedi, S.B., Mineta, K., Gao, X. et al. Computational network analysis of host genetic risk variants of severe COVID-19. Hum Genomics 17, 17 (2023).

Download citation

  • Received:

  • Accepted:

  • Published:

  • DOI: