key: cord-0682274-yj6eo9a8
authors: Lorenzo, Maria A.; Pachón, Diana; Maier, Alexandra; Bermúdez, Henry; Losada, Sandra; Toledo, Marilyan; Pujol, Flor H.; de Noya Belkisyole, Alarcón; Noya, Oscar; Serrano, Maria Luisa
title: Immunoinformatics and Pepscan strategies on the path of a peptide-based serological diagnosis of COVID19
date: 2021-05-13
journal: J Immunol Methods
DOI: 10.1016/j.jim.2021.113071
sha: 91f730e6b5388f3c0c7371bb27a20107078ab4b6
doc_id: 682274
cord_uid: yj6eo9a8

Several diagnostic tools have been developed for clinical and epidemiological assays. RT-PCR and antigen detection tests are more useful for diagnosis of acute disease, while antibody tests allow the estimation of exposure in the population. Currently, there is an urgent need for the development of diagnostic tests for COVID-19 that can be used for large-scale epidemiological sampling. Through a comprehensive strategy, potential 16 mer antigenic peptides suited for antibody-based SARS-CoV-2 diagnosis were identified. A systematic scan of the three structural proteins (S,N and M) and the non-structural proteins (ORFs) present in the SARS-CoV-2 virus was conducted through the combination of immunoinformatic methods, peptide SPOT synthesis and an immunoassay with cellulose-bound peptides (Pepscan). The Pepscan filter paper sheets with synthetic peptides were tested against pools of sera of COVID-19 patients. Antibody recognition showed a strong signal for peptides corresponding to the S, N and M proteins of SARS-CoV-2 virus, but not for the ORFs proteins. The peptides exhibiting higher signal intensity were found in the C-terminal region of the N protein. Several peptides of this region showed strong recognition with all three immunoglobulins in the pools of sera. The differential reactivity observed between the different immunoglobulin isotypes (IgA, IgM and IgG) within different regions of the S and N proteins, can be advantageous for ensuring accurate diagnosis of all infected patients, with different times of exposure to infection. Few peptides of the M protein showed antibody recognition and no recognition was observed for peptides of the ORFs proteins.

Coronavirus disease is the infectious disease associated with the novel coronavirus, named severe acute respiratory syndrome coronavirus 2 (SARS-CoV-2) ("Q&A on coronaviruses ," n.d.). Since its emergence in December 2019 in Wuhan, China, the virus has infected around 100 million people and around 2 million deaths until January 2021("Home -Johns Hopkins Coronavirus Resource Center," 2021). SARS-CoV-2 is the third highly pathogenic coronavirus to pass from an animal reservoir to infect humans in the 21st century: the emergence of recent outbreaks has proved that these viruses can mutate or recombine to become pathogenic and cross the species barriers. Since genetic changes are inevitable and a part of the evolutionary process, these viral outbreaks will keep on emerging (Rabaan et al., 2020) . The high infectivity of SARS-CoV-2 and the worldwide spread of this ongoing outbreak highlight the urgent need for the development of effective diagnostic methods, therapeutics and vaccines. SARS-CoV-2 is a spherical positive single-stranded RNA virus formed primarily of structural proteins currently known as spike (S), membrane (M), envelope (E), and nucleocapsid (N) proteins and a number of non-structural and accessory proteins, whose open reading frames (ORFs) localize between S and For this purpose, a systematic scan of the three structural proteins (S, N and M) and the non-structural proteins (ORFs) present in the SARS-CoV-2virus was conducted through computational methods to predict the potential antigenic regions (B-cell epitopes). Selected regions were synthesized using the SPOT technique and evaluated with the three most important classes of immunoglobulins (IgG, IgM and IgA) by Pepscan. The most antigenic peptides (spots) are the top candidates for the diagnosis of COVID-19.

The reference genome used throughout this study is from the severe acute respiratory syndrome coronavirus 2 isolate Wuhan-Hu-1 (GenBank:

MN908947.3). The protein sequences Spike protein (S) (QHD43416.1), nucleocapsid phosphoprotein (N) (QHR63298.1), membrane glycoprotein (M) (QHD43419.1) and the non-structural proteins ORF1ab (QHD43415),ORF3a (QHD43417.1), ORF6 (QHD43420.1), ORF7a (QHD43421), ORF8 (QHD43422.1) and ORF10 (QHI42199.1) were retrieved from NCBI GenBank database.

Linear B cell epitope predictions were carried out on the spike glycoprotein (S), nucleocapsid phosphoprotein (N), membrane glycoprotein (M) and ORFs proteins (1ab, 3a. 6, 7a, 8 and 10) using several algorithms: Antheprot (standalone) (Deléage, 2012) , SVMTrip (Yao et al., 2012) , ABCPred (Saha and Raghava, 2006) , Kolaskar&Tongaonkar Antigenicity (Kolaskar and Tongaonkar, 1990 ), Bebipred 2.0 (Jespersen et al., 2017) and Ellipro (Ponomarenko et al., J o u r n a l P r e -p r o o f Journal Pre-proof 2008) . The parameters are all set as default.

For prediction of discontinuous or conformational B-cell epitopes, coordinates of experimental structures of SARS-CoV-2 proteins were downloaded from the PDB. In case multiple structures were available, the structure with highest sequence coverage and highest resolution were selected. Conformational Bcell epitopes were predicted with Ellipro (Ponomarenko et al., 2008) In the case of spike protein (S) the crystallized structures were used to locate the 16 mer peptides and to adjust the epitopes. Additional characteristics like surface accessibility and hydrophobicity (GRAVY index < 0) were taken into account to obtain an initial peptide library. The library was enriched with peptides where some mutations were included (Cys→ Ser). Seventy-five peptides were selected to be manually synthesized using the Spot technique.

For N and M proteins an initial library of overlapping peptides was built. Every peptide was 16 amino acid residues in length, offset by four and overlapped by eleven residues, obtaining a total of 135 peptides. Then, the sequences were reduced after elimination of transmembrane and highly hydrophobic peptides (GRAVY index > 0). Also, CBTope (Ansari and Raghava, 2010 ) was used to predict conformational B cell epitopes using primary sequence. Linear and J o u r n a l P r e -p r o o f conformational epitopes where inspected taking into account additional characteristics like Chou Fasman beta-turn prediction (Chou and Fasman, 2006) , Emini Surface Accessibility Prediction (Emini et al., 1985) , and Karplus&Schulz Flexibility Prediction (Karplus and Schulz, 1985) . Additionally, IUPred2A (Abor et al., 2018) was used to identify disordered regions and RADAR (Heger and Holm, 2000) for repetitive regions. Then, consensus regions between the different predictors used were determined. A total of 73 peptides data set was obtained for these proteins.

For the ORF proteins (1ab, 3a, 6, 7a, 8 and 10) the same approach described for the N and M proteins was used. Consensus regions where determined and a final 58 peptides data set was obtained.

The procedure previously described was used with some modifications (Winkler et al., 2009) : The selected peptides were synthesized in spots on filter paper Whatman® grade 540 as a solid support (rectangle 12 x 7 cm), functionalized with a solution of Fmoc-Ala-OH, (0.960 g, 3.0 mmoles), N,N'diisopropylcarbodiimide (DIC) (560 μL, 3.6 mmoles) and N-methylimidazole (475 μL, 6.0 mmoles) in 25 mL of N-methylpirrolidine (NMP) overnight without agitation. Stock solutions of 0.33 M Fmoc-L-aminoacids and 0.5 M of 1hydroxibenzotriazole (HOBt) were prepared in dry NMP. Aliquots of 100 μL of these solutions were activated by the addition 33 μL of a solution 1.1 M DIC in NMP to use during the whole day, after that the aliquots were discarded. The Fmoc group deprotection was achieved using a solution of 20% 4methylpiperidine in NMP and the synthesis was monitored using a 0.02% J o u r n a l P r e -p r o o f bromophenol blue solution in ethanol. The final side chain deprotection was carried out in two steps, first by treatment with a cleavage solution I: 90% trifluoroacetic acid (TFA), 5% water 3%, triisopropylsilane (TIPS), 1% phenol and 1% dichlorometane (DCM) for 30 min; followed by a cleavage solution II: 50% TFA, 2% water, 3% TIS, 1% phenol and 44% DCM for 3h. Both cleavages were performed without agitation. The reagents were from Iris Biotech and Merck Millipore, Germany.

This retrospective study included sera from adult patients and healthy volunteers. Serum samples (n = 10) were obtained from COVID-19 patients confirmed by rapid test (Wondfo® and Avioq, Bio-tech) and/or RT-qPCR (Sansure®) and considered positive if reactive by two independent test. These sera were grouped in two pools of 5 sera each (pool 1 and pool 2). The days since the onset of the symptoms were collected from medical records. All patients were in the acute phase of infection and had a maximum of 30 days of evolution. Control non-SARS-CoV-2 samples (n =5) anonymous stored residual serum samples collected in 2008 were used.

This research project was approved by the Independent Bioethics Committee for Research of the National Center for Bioethics (CIBI-CENABI, Venezuela). Serum samples were provided by patients who accepted to participate in this study and signed the Informed Consent.

The dry paper with the different "spots" of peptides was blocked with phosphate buffer saline, pH 7.4 with 0.05% Tween 20 (PBS-Tween) and 5% non-fat dry J o u r n a l P r e -p r o o f milk and tested with two pools of five patients' sera with a final dilution of 1:200.

Anti-Human Polyvalent Immunoglobulins (G,A,M), Anti-human IgG, anti-human IgA and anti-human IgM-horseradish-peroxidase (Sigma) were used at 1:30.000, 1:30.000, 1:3.000 and 1:10.000, respectively, as conjugates and revealed with a solution of SuperSignal™ West Pico PLUS. The chemiluminescence of the spots was measured using the ChemiDoc™ Imaging Systems equipment and the images obtained were processed with the ImageLab program (BioRad).

The removal of bound proteins was carried out in two steps. In the first step the membranes were treated with the regeneration buffer I (solution 8 M of urea with 1% SDS, 3 times for 10 minutes). Then, they were treated with regeneration buffer II (solution of 100 mL of AcOH (99%), 500 mL of EtOH and 400 mL of water, 3 times for 10 min) both at room temperature. Finally, they were washed at least five times with EtOH and dried on air. At this point, the membranes were used immediately or stored at -80 ºC until use. The regeneration was checked blocking the membranes and incubating with the corresponding enzyme-conjugate to ensure no spots remained. The regeneration process was carried out 3 times after the evaluation of antigenicity.

In some cases, when it was not successful, the regeneration times using buffer I were increased from 1 hour to 24 hours.

In the present study, our aim was the identification of potential antigenic peptides well suited for antibody-based SARS-CoV-2 diagnosis. First, we J o u r n a l P r e -p r o o f developed an immunoinformatic strategy to obtain a 16 mer peptides data set of 206 peptides of proteins S, N, M and ORFs.

To experimentally determine the antigenicity of the selected peptides, the 206 sequences were synthesized on functionalized cellulose papers in a spot manner. Three membranes were prepared using the SPOT technique and identified as 1, 2 and 3. Each membrane contains 76 peptides (spots) covalently anchored to the cellulose paper, which was previously functionalized with the amino acid residue -alanine. Additionally, a spacer was added using the same amino acid for C-terminus functionalization. Each membrane presents spots repeats as positive controls. All synthesized peptides were acetylated at the N-terminus to eliminate the non-physiological positive charge that would otherwise be present on the α-amino group of the N-terminal residue of an internal peptide sequence (Carter and Loomis-Price, 2004) . Figure 1A shows the distribution of the peptides (spots) of the proteins studied in each membrane. Membranes were exposed to two different pools (pool 1 and pool 2) of five COVID-19 patients´ sera that were highly reactive, moderately reactive and slightly reactive to Wondfo® y Avioq rapids test (Table 1) 194 +++ f+ + + "+" positive result; number of "+" denotes relative increase in positive band intensity in the strip tests. "f+" stands for faint-positive. "-" negative result. "nd" not determined.

Membranes were evaluated for the three most important classes of immunoglobulins (IgG, IgA and IgM). A regeneration process was carried out and checked for each membrane before using different immunoglobulins, to ensure the removal of primary and secondary antibodies, and thus avoiding false positives during the assays.

When exposed to pools of positive COVID-19 sera, antibody recognition showed a strong signal for peptides corresponding to the S, N and M proteins of SARS-CoV-2 virus, but not for the ORFs proteins. Figure 1C shows peptides reactivity against pools of sera (since no reactivity to ORFs peptides was observed, membrane 3 is not shown). This methodology was successfully used and allowed us to quickly identify antigens based on synthetic peptides. In general, higher IgA and IgM and reactivity was observed, compared to the IgG one. Regardless of the pool used, a higher number of peptides were recognized J o u r n a l P r e -p r o o f by IgM immunoglobulin, followed by IgA recognition. Few peptides (spots) were recognized by IgG, as expected. The serum samples used in the present study were collected from patients with COVID-19 at early stages of the viral infection (acute-phase) when IgM and IgA immunoglobulins are abundant, while IgG is the less abundant circulating antibody in the initial phase of infection. It has been reported that the level of IgG antibody reached the highest concentration on day 30, while the highest concentration of IgM antibody appeared on day 18 but then began to decline (Hueston et al., 2020; Qu et al., 2020) . This might be one of the reasons for this differences in the isotypic responsiveness to the different peptides. The results obtained for each of the aforementioned proteins is described below. The increasing reactivity of the peptides in the Pepscan spots, rises from gray to black and when it is very intense it turns red. In each block the upper pair shows the results for sera pool 1(P1) and the lower for pool 2 (P2). Reactivity (absent) to peptides included in Membrane 3 is not shown.

Membranes were also exposed to a pool of five non-COVID-19 patients´ sera, that were collected in 2008. For these assays a commercial mixture of total anti-immunoglobulins was used. No significant antibody recognition was observed ( Figure S1 ).

The spike glycoprotein (S) forms homotrimers protruding the viral surface and contains a S1 subunit, comprising the receptor binding domain (RBD) that mediates the binding to host cell ACE2 receptor, and the S2 subunit, which contains the fusion peptide and functions in membrane fusion. Spike protein has become the major target of antibodies responses, particularly the RBD domain (Barnes et al., 2020; Walls et al., 2020; Wrapp et al., 2020) .

To select candidate diagnostic peptides, we performed predictions for linear B cell epitopes with Bepipred 2.0 and Ellipro, and for conformational epitopes with Ellipro and Discotope 2.0. To predict and map conformational B cell epitopes, we used the SARS-CoV-2 spike glycoprotein structures (PDB: 6VSB, 6MOJ).

Several physicochemical characteristics, such as gravy index and hydrophobicity, were also considered.

Then, the relevant amino acid positions were localized onto the model structures. The X-ray structure of the RBD domain of the S protein (PDB 6M0J) allowed us to map the epitopes and to determine the final 16 mer sequences for peptide synthesis (Figure 2A, 2B) . The x-ray structure of the S protein spike in the prefusion conformation (PDB 6VSB) was used to map the predicted epitopes and to determine the sequences of the peptides to be synthesized by the spot technique ( Figure This analysis allowed us to select peptides that are clearly exposed. Three of the regions selected are located in the S1 subunit in the NTD and RBD domains, whereas others are in the S2 subunit.

Additionally, we used the SARS-CoV spike protein structure (PDB:6CRZ) to predict epitopes and to select the sequences to be synthesized (Data not shown). To determine the antigenicity of the selected peptides of the S protein, these were synthesized on functionalized cellulose papers in a spot manner ( Figure   1A ) and exposed to two different pools of COVID-19 patients´ sera, as previously described. Antibody recognition was observed for several spots ( Figure 1C, M1 and Figure 3 ). The sequences with higher signal intensity are localized in domains S1 A (spot 5), RBD (spots 39, 41, 43, 49, 52, 55, 62; spot 62 was added as repetition of 55 for chemical control), and S1 C (spots 17, 21, 22,23, 54, 61; spot 61 was added as repetition of 54 for chemical control ), within the S1 subunit. At the S2 subunit, three sequences showed strong signal intensity (spot 32, 34 and 37). Other spots showed slight antibody recognition.

Serum antibody responses against the peptides (spots) with higher signal intensity can be characterized in more detail. Several peptides (spots 5, 17, 18, 20, 21, 22, 32, 34, 39, 41, 43, 49, 52, 54 and 62 . Figure 1C , M1 and Figure 3) were strongly recognized by IgM and IgA immunoglobulins. Those are localized along the different antigenic regions at the S1 and S2 subunits and RBD domain. Peptides 5 and 17 (localized in domains S1 A and S1 C , respectively) showed strong IgM and IgA recognition with the two pools of sera. Peptides 43, 49 and 62, localized in the RBD domain, only showed a strong recognition with pool 2. Peptides 22, 34, 39, 41 and 52, showed recognition by the IgM immunoglobulin with both pool´s of COVID-19 patients´ sera. These peptides are localized mainly in the RBD domain of the spike protein, with the exception of peptides 22 and 34, found in the S1 C and S2 subunit, respectively. Among all evaluated peptides, IgG recognition was achieved by peptide 23, located at the S1 C domain and peptide 37, located at de S2 domain. Spot 23 showed a strong signal when exposed to pool 1 of sera but a slight antibody recognition Our analysis also considered several sequences of the SARS-CoV spike protein (Table 2 ). With the exception of spot 18 and 20, when those peptides were exposed to the two pools of COVID-19 patients´ sera, no significant antibody recognition was observed. 

Region IgG IgM IgA 5 S1 A -/ + +++ / +++ + / ++ 17 S1 C -/ +++ +++ / +++ -/ +++ 18* S1 C ++ / ++ + / ++ +++ / ++ 20* S1 C -/ --/ -++ / -21 S1 C +++ /+++ + / +++ +++ / +++ 22 S1 C -/ -++ / ++ -/ -23 S1 C +++ / --/ --/ - 

Protein N(McBride et al., 2014) is frequently used structural protein for immunodiagnostic of SARS-CoV-2, it is produced at very high levels in the virus infected cells and is considered a good candidate for clinic diagnosis . The N protein is a highly immunogenic, antigenic and abundantly Figure 5 ). These spots showed antibody recognition with different signal intensities when exposed with the two pool of sera. Nevertheless, peptide 52 showed anti-IgG positive strong recognition only with the pool 2 of sera, IgA recognition with the two pool of sera but no IgM recognition ( Figure 5 and Table   3 ). Peptides of this region were highly hydrophilic (GRAVY index of -2.2) when compared with the other peptides synthesized in membrane 2.

J o u r n a l P r e -p r o o f C-terminal + / ++ -/ -f+ / +++ Results Pool 1/Pool 2. "+" denotes positive result; number of "+" signifies relative increase in positive spot intensity in the membrane; "f+" stands for faintpositive. "-" denote negative result.

Other regions outside the C-terminal fragment of the N protein exhibited strong IgG, IgM and IgA recognition ( Figure 5 and Table 3 ). One of these regions are localized in the SR-rich linker and CDT regions (spots 26, (30) (31) (32) (38) (39) Figure1C, M2) , and showed different immunoreactive response with the two pools of sera and with the all three IgG, IgM and IgA immunoglobulins ( Figure   5 ). In Table 3 it can be clearly observed there is no IgG recognition for any of the peptides, regardless of the pool evaluated, except for peptide 31 that showed IgG recognition. However, these peptides were strongly recognized by IgM immunoglobulin with the two pool of sera. These regions are less hydrophilic that the C-terminal fragment of the N protein, according to the GRAVY index analysis and could support the relationship between antigenic response and hydrophilicity (data not shown). Antibody recognition was also observed for several spots at the N-terminal region (spots 1, 2 and 16), where no specific recognition pattern was observed.

To date, there are very few reports about the antigenic properties of the transmembrane M glycoprotein and its importance in the diagnosis of COVID-J o u r n a l P r e -p r o o f 19. Both SARS-CoV-2 and SARS-CoV are closely related, belonging to the βcoronavirus genus, and the M protein function is expected to be similar.

The M protein (also known as E1 membrane glycoprotein or matrix protein) is one of major membrane proteins of coronaviruses, together with the S (spike) and the E (envelope) proteins. The M protein contains three domains: a short N terminal ectodomain, a triple-spanning transmembrane domain, and a Cterminal endodomain (Armstrong et al., 1984) . It has been postulated that the M protein is related to viral infectivity through binding to the viral S protein and the host surface receptor(s), thus promoting membrane fusion. It appears to be involved in the antigenicity demonstrated by the virus-induced immune responses of the host (Hu et al., 2003) .

For diagnostic peptide selection, a similar strategy that the one applied to protein N, was followed. To date, the crystal structure of the SARS-CoV-2 M protein have not been reported. To map the predicted epitopes and to design the peptides finally synthesized, the homology model from the I-Tasser server was used (code QHD43419)("Modeling of the SARS-COV-2 Genome using I-TASSER," n.d.) (Figure 6 ). (Table 4) . It is worth to mention that pool 1 contained highly reactive of COVID-19

patients´ sera (2 of 5), reactive (2 of 5) and one low reactive (1 of 5). On the other hand, pool 2 consisted of low reactive sera (3 of 5), one high reactive and one moderately reactive. The results obtained using pool 1 showed that the immunoglobulins (IgA, IgM and IgG) recognition by the different spots were similar. When the peptides were exposed to pool 2, IgA mild recognition was observed for a greater number of spots ( Figure 1C , M2 and Figure 5 ).

None of the evaluated peptides were able to discriminate between the acute or chronic infection. Nevertheless, to make this statement, additional evaluation is required with individual COVID-19 patients´s sera. The heterogeneous recognition of the peptides by different immunoglobulins of the patients whose appearance is determined by the time of evolution of the infection, indicates that before the development of a diagnostic kit, polyvalent conjugates capable of capturing the three main immunoglobulins should be used.

Although there are few reports on the usefulness of ORF proteins in diagnosis, 

Most of the peptides evaluated from these proteins were weakly recognized, when compared with the structural proteins S, N and M. Of all the non-structural proteins tested, only peptides derived from the ORF8 and ORF10 proteins were recognized by any of the different immunoglobulins ( Figure 1C , M1, Table 5 and Figure 3 ). Table 5 it can be clearly observed that the selected peptides, localized in the C-terminal region of the ORF10 protein, showed recognition by the IgG immunoglobulin only with pool 1 of COVID-19 patients´ sera and that peptides 66, 99 and 70, showed recognition by the IgA immunoglobulin but only with pool 2 of sera. One peptide (spot 70) was recognized by the IgM immunoglobulin with the two pools of sera.

The ORF8 protein is a small protein with only 121 amino acid residues, represented by peptides (spots 72 to 76) of the N-terminal region. A recent work show that the combinational use of ORF3b and ORF8 tests alone could be sufficient to detect individuals exposed to COVID-19 at any time point of infection (Hachim et al., 2020) . However, in our case, at least with the evaluated peptides, only a moderate IgA recognition was observed with pool 2 of COVID-19 patients´ sera.

In this study, we use a comprehensive strategy to identify potential 16 mer antigenic peptides suited for antibody-based SARS-CoV-2 diagnosis, in order to be used with different immunological techniques. A systematic scan of the three structural proteins (S, N and M) and the non-structural proteins (ORFs) present in the SARS-CoV-2 virus was conducted through the combination of immunoinformatic methods, peptide SPOT synthesis and immunoassay by cellulose-bound peptides (Pepscan). The membranes with synthetic peptides were tested against two pools of sera from COVID-19 acute infected patients. A strong antibody recognition was observed for peptides corresponding to the S, N and M proteins of SARS-CoV-2 virus, but not to the ORFs proteins. IgM and IgA immunoglobulins highly recognized several peptides of the S protein.

Those are localized along the different antigenic regions at the S1 and S2 subunits and RBD domain. The sequences with higher signal intensity were 

IUPred2A: contextdependent prediction of protein disorder as a function of redox state and protein binding B ´ alint M ´ eszárosesz´eszáros

Identification of conformational B-cell Epitopes in an antigen from its primary sequence

Sequence and topology of a model intracellular membrane protein, E1 glycoprotein, from a coronavirus

Insights into the recent 2019 novel coronavirus (Sars-coV-2) in light of past human coronavirus outbreaks

Structures of Human Antibodies Bound to SARS-CoV-2 Spike Reveal Common Epitopes and Recurrent Features of Antibodies ll Structures of Human Antibodies Bound to SARS-CoV-2 Spike Reveal Common Epitopes and Recurrent Features of Antibodies

B Cell Epitope Mapping Using Synthetic Peptides

Advances of bioinformatics tools applied in virus epitopes prediction

Prediction of the Secondary Structure of Proteins from their Amino Acid Sequence

An interactive 3D viewer of molecules compatible with the suite of ANTHEPROT programs

Induction of hepatitis A virus-neutralizing antibody by a virus-specific synthetic peptide

Genomic characterization of the 2019 novel humanpathogenic coronavirus isolated from a patient with atypical pneumonia after visiting Wuhan

Synthetic Peptides for the Immunodiagnosis of Human Diseases

The explosion of new coronavirus tests that could help to end the pandemic

ORF8 and ORF3b antibodies are accurate serological markers of early and late SARS-CoV-2 infection

Rapid automatic detection and alignment of repeats in protein sequences

Hopkins Coronavirus Resource Center

The M protein of SARS-CoV: basic structural and immunological properties

The Antibody Response to SARS-CoV-2 Infection

BepiPred-2.0: improving sequence-based B-cell epitope prediction using conformational epitopes

Crystal structure of SARS-CoV-2 nucleocapsid protein RNA binding domain reveals potential unique drug targeting sites

Prediction of chain flexibility in proteins -A tool for the selection of peptide antigens

A semi-empirical method for prediction of antigenic determinants on protein antigens

False Positive Results Of IgM/IgG antibodies against antigen of the SARS-CoV-2 in sera stored before the 2020 Endemia in Italy

Generation of antibodies against COVID-19 virus for development of diagnostic tools

Systematic evaluation of IgG responses to SARS-CoV-2 spike protein-derived peptides for monitoring COVID-19 patients

The coronavirus nucleocapsid is a multifunctional protein

Modeling of the SARS-COV-2 Genome using I-TASSER [WWW Document

SARS-CoV-2 epitope mapping on microarrays highlights strong immune-response to n protein region

Immunodiagnosis of Parasitic Diseases with Synthetic Peptides

ElliPro: A new structure-based tool for the prediction of antibody epitopes

Q&A on coronaviruses

Profile of Immunoglobulin G and IgM Antibodies Against Severe Acute Respiratory Syndrome Coronavirus 2 (SARS-CoV-2)

SARS-CoV-2, SARS-CoV, and MERS-CoV: a comparative overview

Prediction of continuous B-cell epitopes in an antigen using recurrent neural network

Immunoinformatics may lead to a reappraisal of the nature of B cell epitopes and of the feasibility of synthetic peptide vaccines

Structure, Function, and Antigenicity of the SARS-CoV-2 Spike Glycoprotein

SARS-CoV-2 Proteome Microarray for Mapping COVID-19 Antibody Interactions at Amino Acid Resolution

Synthesis of peptide arrays using SPOT-technology and the CelluSpots-method

Cryo-EM structure of the 2019-nCoV spike in the prefusion conformation

SVMTriP: A Method to Predict Antigenic Epitopes Using Support Vector Machine to Integrate Tri-Peptide Similarity and Propensity

Subcellular localization and membrane association of SARS-CoV 3a protein

COVID-19 diagnosis -A review of current methods