key: cord-0939754-cj4qd3yw
authors: Ren, Lili; Yang, Donghong; Ren, Xianwen; Li, Mingkun; Mu, Xinlin; Wang, Qi; Cao, Jie; Hu, Ke; Yan, Chunliang; Fan, Hongwei; Li, Xiangxin; Chen, Yusheng; Wang, Ruiqin; An, Fucheng; An, Shuchang; Luo, Ming; Wang, Ying; Xiao, Yan; Xiang, Zichun; Xiao, Yan; Li, Li; Huang, Fang; Jin, Qi; Gao, Zhancheng; Wang, Jianwei
title: Genotyping of human rhinovirus in adult patients with acute respiratory infections identified predominant infections of genotype A21
date: 2017-01-27
journal: Sci Rep
DOI: 10.1038/srep41601
sha: 59f96c4166966aa929450a859c3d3fb2b6058c50
doc_id: 939754
cord_uid: cj4qd3yw

Human rhinovirus (HRV) is an important causative agent of acute respiratory tract infections (ARTIs). The roles of specific HRV genotypes in patients suffering from ARTIs have not been well established. We recruited 147 adult inpatients with community-acquired pneumonia (CAP) and 291 adult outpatients with upper ARTIs (URTIs). Respiratory pathogens were screened via PCR assays. HRV was detected in 42 patients, with 35 species A, five B and two C. Seventeen genotypes were identified, and HRV-A21 ranked the highest (9/42, 21.4%). The HRV-A21-positive infections were detected in four patients with CAP and in five with URTIs, all without co-infections. The HRV-A21 genome sequenced in this study contained 12 novel coding polymorphisms in viral protein (VP) 1, VP2 EF loop, VP3 knob and 3D regions. The infections of HRV-A21 virus obtained in this study could not be neutralized by antiserum of HRV-A21 prototype strain (VR-1131), indicating remarkable antigenic variation. Metagenomic analysis showed the HRV-A21 reads were dominant in bronchoalveolar lavage fluid of the three HRV-A21-positive patients with severe CAP, in which two dead. Our results highlight an unexpected infection of genotype HRV-A21 in the clinic, indicating the necessity of precise genotyping and surveillance of HRVs to improve the clinical management of ARTIs.

Genotyping of human rhinoviruses in ARTI patients. To identify the roles of HRV genotypes in adult ARTIs, we recruited 438 patients from January to December 2013, including 147 community-acquired pneumonia (CAP) inpatients (including 39 severe cases) and 291 Upper ARTI (URTI) outpatients. The CAP inpatients ranged from 18 to 92 years old, with a median age of 55.5 [Interquartile range (IQR) years. The URTI outpatients ranged from 18 to 84 years old, with a median age of 30 (IQR 24-41) years.

A total of 169 (38.6%) patients tested positive for at least one respiratory pathogen via multiplex real-time PCR. Forty-two (9.6%) patients were positive for HRV, including 15 CAP and 27 URTI patients (Table 1) . Co-detection was found in ten HRV-positive cases, three of which were CAP patients (co-detection with Mycoplasma pneumoniae, Streptococcus pneumoniae, or HCoV-229E, respectively). The other seven cases were URTI patients, four of whom were also infected with influenza viruses (IFVs), two with respiratory syncytial virus (RSV), and one with EV. HRV RNA was not detected in the sera of HRV-positive patients, as determined by RT-PCR. The HRV species for these patients were as follows: 35 HRV-A, 5 HRV-B, and 2 HRV-C. Among them, 29 HRV-A and 5 HRV-B were successfully genotyped.

In the CAP patients, there were four cases of HRV-A21 and two of A68. The rest of the genotypes were one case each of A2, A9, A13, A33, A53, B35, B48, B84, and C un-typed. Three of the four HRV-A21-positive CAP inpatients had serially collected respiratory samples, and all were positive for HRV-A21. In the URTI patients, there were five HRV-A21, three A61, two A98, and one case of each of the following genotypes: A2, A28, A53, A58, A60, A68, A96, B27, B84, and C un-typed. Seven samples of HRV-A in the URTI patients could not be genotyped precisely ( Table 1 ). The detection rate of HRV-A21 was higher in the severe CAP patients (7.9%, n = 3) than in the non-severe CAP (0.9%, n = 1) or the URTI patients (1.7%, n = 5) (P = 0.047 by Fisher's exact test). Taken together, these data show that prevalence of HRV-A21 correlates with a spectrum of illness from mild URTIs to severe CAP. Unexpectedly, HRV-A21 was most frequency (9 of 42, 21.4%) detected as the only pathogen present in an infection among the identified HRV genotypes. As only three genome sequences of HRV-A21 were available in GenBank and little was reported on its clinical manifestation, we focused on the characterization of A21 infections.

Complete genomic sequences were obtained from all nine HRV-A21-positive cases. Phylogenetic analysis showed that the HRV-A21 strains identified in this study were closely related to the three available reference sequences from GenBank, including the prototype strain (ATCC VR-1131; FJ445121) and two strains identified in the USA in 2000 (p1295_s3993_2000; JQ747747) and 2010 (p1177_sR3307_2010; JN837693) ( Fig. 1A) , but they formed a distinct clade in the phylogenetic trees of complete viral genome and VP1 gene), indicating the presence of variants of the newly isolated HRV-A21 strains. Multiple alignment analyses among the HRV-A21 strains in this study showed that the nucleotide (nt) identities of the genomes were 99.0-99.7%, whereas the amino acid (aa) sequence identities of VP1 were 98.6-100.0%. Using the RMH001/2013 strain as a representative, the nt and aa sequences of VP1 compared to the reference sequences were 90.4-92.1% and 94.7-96.5%, respectively ( Table 2 ). The genetic diversity of the newly identified HRV-A21 was further characterized by aa alignments. A total of 12 specific aa substitutions in the polyproteins of the HRV-A21 strains in this study were observed, including three changes (L/Q234E, K235 R, and A/S241T) in the puff region (EF loop) of VP2, one substitution (T389A) in the VP3 "knob, " five substitutions (Q798E, V816I, N/D 841E, K/E 849R, and R852K) in VP1, and three substitutions (R1743K, M1767I, and K2061Q) in the 3D region (reference strain RMH001/2013KM576764). The 241T substitution was only found in the HRV-A21 strains in this study but not in any of the three HRV reference sequences. No specific substitutions were found in the BC or DE loops of the VP1 region and other coding gene regions (Fig. 1B) .

To assess the genetic variation of the HRV-A21 strains identified in this study, we then conducted a neutralization assay using a HRV-A21 strain isolated in this study and the antiserum against the prototype strain (ATCC VR-1131). The results showed that the antiserum against prototype HRV-A21 could not neutralize the infections of the new isolate, indicating that the mutations in the viral genome represent remarkable antigenic variations.

Clinical manifestations of HRV-A21-positive patients. We then characterized the manifestation of HRV-A21-positive patients. The signs and symptoms were found to be similar to those of patients who were positive for other HRV genotypes and other respiratory viruses upon hospital admission. The median age of HRV-A21-positive CAP patients were older than that of the five URTI outpatients (70 years vs. 24 years, P = 0.037 by Mann-Whitney U test) ( Table 3 ). Three CAP patients and two URTI patients positive for HRV-A21 had leukocytosis with neutrophilia, and others showed leukopenia at presentation. The viral loads of HRV-A21 in the respiratory samples ranged from 3.7 × 10 4 to 2.8 × 10 8 copies/ml, with a median of 2.4 × 10 6 (IQR 1.3 × 10 6 -2.1 × 10 7 ) copies/ml, and no significant differences were observed between the CAP inpatientsand outpatients (1.3 × 10 6 vs. 1.95 × 10 7 copies/ml, P = 0.19 by Mann-Whitney U test).

Three of the four HRV-A21-positive CAP patients (coding RMH001, RMH114, and RMH123) deteriorated rapidly after hospitalization and developed several complications, including respiratory failure, septic shock, and acute renal failure. They were then admitted to the intensive care unit and supported with mechanical ventilation. Two of these three patients (coding RMH123 and RMH114) died 26 and 23 days after the onset of illness, respectively. Both fatal cases had underlying diseases; one had cerebrovascular disease and the other had Parkinson's disease. The other HRV-A21-positive patients recovered.

To determine whether other microbes present as co-pathogens may be related to the observed severity of the HRV-A21 infections (RMH123, RMH114, and RMH001), we carried out metagenomic analysis using the serially collected bronchoalveolar lavage fluid (BALF) or tracheal aspirate samples. HRV-A21 reads were overwhelmingly predominant in the tested samples of the microbial reads, representing 68.7% and 44.8% on days 4 and 9, respectively, in RMH123; 99.9% of the reads on days 12 and 14 in RMH114; and 95.0% and 86.0% of the reads on days 3 and 5 in RMH001 after the onset of symptoms (Fig. 2) .

Deep sequencing of the lower respiratory samples from the three severe CAP patients identified 40 single nucleotide polymorphisms (SNPs), including 24 nonsynonymous mutations. These SNPs were located in the VP1, VP2, VP3, 3C, and 3D gene regions (see Supplementary Table S1 ). Two SNPs switched between minor and major alleles in serial RMH123 viral strains, 3135 (aa 841 in VP1 from E to K), and 5740 (aa 1709 in 3D from C to Y) (reference strain RMH001/2013KM576764). The aa site E841 was the specific site mutation in all the HRV-A21 strains identified in this study. No specific site mutations were found in sequences obtained from RMH123 and RMH144, the two dead cases.

Reads that mapped to known pathogens were also found but at lower level. Sequence data corresponding to Acinetobacter baumannii (23.9% and 13.3%) and human herpesvirus (HSV)-1 (6.5% and 0.4%) of the microbial species were detected in RMH123 on days 4 and 6, respectively. In RMH001, on day 3 after the onset of symptoms, reads corresponding to Escherichia coli, Haemophilus influenza, and Streptococcus pneumoniae accounted for 1.2%, 1.2%, and 0.8% of the identified microbial species, respectively, in BALF. GB virus C, HSV-1, and hepatitis B virus reads were found on day 5 after the onset of symptoms and accounted for 8.8%, 3.0%, and 1.3% of the microbial species, respectively, in the tracheal aspirate.

In this study, we examined the HRV genotypes presented in ARTIs. Most of the identified genotypes were sporadically detected. Unexpectedly, we found that HRV-A21, a clinically rarely reported HRV genotype in literatures before, appears to be present in a range of illnesses from mild URTIs to severe CAP. Moreover, HRV-A21 was the most frequently detected strain among all HRV genotypes detected. HRV-A21 was more frequently detected in patients with severe CAP than in those with non-severe CAP and URTIs. Further analysis showed that the HRV-A21 detected in this study was distinct from the only three strains of HRV previously identified by phylogenic analysis, as demonstrated by the neutralization assay, during which one isolate obtained in this study could not be neutralized by the antiserum against the prototype strain. Metagenomic assay showed that infection with HRV-A21 is strongly associated with sporadic severe CAP in adults.

Respiratory infections with HRV-A21 were scarcely recognized clinically prior to this study. Only three whole HRV-A21 genome sequences were available in GenBank: ATCC VR-1131 (prototype strain), and two sequences obtained directly from clinical patients collected in 2000 and 2010. The only other case of HRV-A21-associated severe respiratory disease was reported in France in 2010, in which one child with pneumonia tested positive for HRV-A21, but only partial VP4/VP2 gene sequences (420 bp) were deposit in GenBank (accession number HE589784) 11 . Our findings on the clinical manifestation and virological features of this genotype provide insights into the pathogenesis of HRVs and new knowledge on the roles of different HRV genotypes in ARTIs, which is informative and necessary to improve the management of ARTIs.

To investigate the association between HRV-A21 and ARTIs, we conducted metagenomics analysis in addition to PCR. For this purpose, we used only lower, not upper, respiratory samples to avoid the interference of mass microbial species, which can be carried in the upper respiratory tract. The data demonstrated that HRV-A21 reads were predominantly present in the BALF or tracheal aspirate samples collected from three patients with severe CAP. Although other agents, including Acinetobacter baumannii, Escherichia coli, Haemophilus influenza, and Streptococcus pneumoniae, were identified by deep sequencing, they only accounted for a minor part of the microbial reads in early samples, indicating that colonization or possible secondary infections occurred but that they were not the primary causative agents. The HSV-1 reads were identified in the respiratory samples from two patients. However, when compared to those of HRV-A21, they were not the dominant species. It may accounts for the activation of latent viral infections during severe infections. Reads of hepatitis B virus detected in RMH001 in the second respiratory samples might result from the alveolar hemorrhagic effusion. However, the role of these infections in the severity of pneumonia should be intensively investigated in future studies. Taken together, these data suggest that there is an association between HRV-A21 and CAP. In the past, HRV infections were thought to cause changes in the airway microbiome and increase lung inflammation in COPD patients 12 . Whether HRV infection contributes to secondary bacterial infections, such as Streptococcus pneumoniae, and whether HRV is a co-pathogen with these agents during the disease process (as is influenza virus) warrants further investigation 13 . Viral factors and the host immune status are responsible for disease severity during viral infections 13, 14 . The HRV-A21-positive patients in this study were not immunocompromised, but two of the three severe cases had underlying diseases; thus, the factors that contribute to disease severity require further investigation. Although the number of HRV-A21-positive cases was small, senior age may be a risk factor for the degree of CAP severity in this cohort. Similar to the infections with A/H1N1 2009pdm, H7N9 avian influenza virus, and MERS-CoV, different medical settings are also risk factors for disease severity [15] [16] [17] . Genomic variation may also be a determinant of viral adaption and pathogenesis. In the Enterovirus genus, variation can lead to the emergence of new strains or genotypes in epidemics 4, 18 . For example, the emergent EV-D68 that circulated in the USA in 2014 (http://www.cdc.gov/non-polio-enterovirus/outbreaks/EV-D68-outbreaks. html) contained mutant genomic sequences, which could have been responsible for the outbreak and increased disease severity 4, 19 . The VP1 BC and DE loops, the VP2 EF loop, and the VP3 knob regions are known to form the major neutralization sites in picornaviruses and the aa mutations in these regions might change the ability to neutralize the virus 20 . The fact that infections of the new HRV A21 isolate could not be neutralized by the antiserum against the prototype strain indicates that the new HRV-A21 strains are antigenic distinct to the prototype strain. Previous studies have also demonstrated that the single aa site mutation in VP1, the VP2 EF loop, and the VP3 knob enhance the virulence of CVB2, B3, and EV71, and the mutation site in the 3D gene, which encodes the RNA-dependent RNA polymerase of EVs, decreased the virulence of EV71 in mice [20] [21] [22] . In this study, specific mutations in the VP1, VP2 EF loop, VP3 knob, and 3D regions of HRV-A21 strains were identified by deep sequencing and phylogenetic analysis. However, no specific site mutations were found in sequences obtained from HRV-A21-positive dead cases. The A389 mutation in the VP3 knob of HRV-A21 identified in this study, corresponding to the site in CVB3 has been demonstrated relating to the virulence in CVB3 21 . Whether the others HRV-A21 mutations facilitate host adaption and replication, especially in the lower respiratory tract tropism, and result in clinical disease phenotypes remains unclear and warrants further investigation. We also found the switches between minor and major alleles in serially collected HRV strains of RMH123, a dead case, which indicates the occurrence of viral quasispecies during infections. The fact that several HRV variants co-circulate in the population and that we do not have waves of one unique virus indicates a need for precise genotyping of HRV infections. Our findings highlight that HRV should not be underestimated in severe respiratory illness.

Metagenomic analysis has been widely used for viral identification, sequencing, characterization, and origin tracing 23 . The implementation of such methodologies has greatly improved the identification of pathogens, especially for mutant strains and novel viruses. For example, deep sequencing data has been used as direct evidence to identify an H10N8 influenza virus infection in a single human case 24 . Optimization and implementation of the protocols suitable for clinical samples will no doubt improve the microbial diagnosis in clinical practice.

In summary, we identified the predominant infection of HRV-A21 in ARTIs that may be related to life-threatening CAP cases. The findings indicate that there still exists much more to establish regarding the pathogenesis of HRV compared to what we have outlined. Intensive surveillance and precise evaluation of specific HRVs in ARTIs is necessary.

Patients and clinical data. URTI and CAP patients between January 1 and December 31, 2013 were included in this study. URTI patients were recruited by the national respiratory virus surveillance project. The inclusion criteria for URTIs were as follows: at least 18 years of age, acute fever (body temperature ≥ 38.0 °C), and respiratory symptoms without radiological pulmonary abnormalities. Acute phase sera and nasal and throat swabs were collected from each outpatient when they sought medical care.

Inpatients diagnosed with CAP were recruited through the China Severe Acute Respiratory Infections network (China SARInet). CAP was diagnosed following the guidelines of the Infectious Diseases Society of America and the American Thoracic Society; criteria included age of at least 18 years, with respiratory symptoms, chest roentgenogram, and/or computed tomography confirmed pneumonia 25 . Severe CAP was diagnosed according to published criteria 26 . Pneumonia patients with confirmed non-inflammatory pulmonary disease were excluded. BALF or tracheal aspirate specimens and sera were collected from each inpatient when clinically indicated. BALF or tracheal aspirate specimens were collected as second respiratory samples in patients with mechanical ventilation, followed by serial sputum collection. The clinical information for each enrolled patient was recorded using a standard form.

The study was approved by the Medical Ethics Review Board of the Institute of Pathogen Biology, Chinese Academy of Medical Sciences. All the methods were carried out in accordance with relevant guidelines and regulations, including any relevant details. Written informed consent for scientific assessment was obtained from each patient as part of the clinical treatment contract.

Total nucleic acids were extracted from respiratory and blood samples using HRV genotyping, load quantification, and phylogenic analysis. Conventional reverse transcriptase PCR (RT-PCR) targeting the 5′ -untranslated region (UTR)/viral protein (VP) gene 4 and the VP4/VP2 region was used for HRV screening and genotyping as described elsewhere [27] [28] [29] . Briefly, the PCR products were sequenced directly using an ABI 3730xl automatic DNA analyzer (Life Technologies, Grand Island, NY, USA). Genotypes Scientific RepoRts | 7:41601 | DOI: 10.1038/srep41601 were determined by phylogenic analysis using the amplified gene sequences compared to the reference HRV genotype sequence deposited in GenBank and by MEGA 5.1 using the neighbor-joining method with Kimura's two-parameter model and 1,000 bootstrap pseudo-replicates 30 . The HRV viral loads in the respiratory samples were determined using an HRV 5′ -UTR quantitative real-time RT-PCR assay and standard curve analysis. The standard RNA transcript was amplified in vitro from the plasmid containing gene fragments using primers targeting UTR/VP4 29 .

Metagenomic analysis. The lower respiratory samples collected from severe CAP patients were used to construct the cDNA libraries and sequenced using an Ion Torrent platform with a type 318 chip (Thermo Fisher Scientific Inc, USA). The samples were treated with Turbo DNase (Life Technologies, USA) and RNAse One (Promega, USA) to decrease the host genome background according to the manufacturer's instructions. Total nucleic acids were extracted and amplified using the Ovation RNA-Seq System (Nugen, California, USA). The libraries for deep sequencing were constructed using the NEBNext Fast DNA Library Prep set for the Ion Torrent (New England Biolabs). Alignments with mapping quality scores less than 30 and reads containing fewer than 50 nucleotides were excluded from the following analysis. The taxonomic assignment was performed using MEGAN software (version 5.2.3) with the parameters "Min Support 50, Min score 80, Top percent 10" 31 . All reads assigned to the Enterovirus genus were retrieved and aligned to the consensus sequence of the reference strain, HRV-A21_ p1177_sR3307_2010 (GenBank accession number JN837693), using BWA (version 0.7.5, BWA-MEM algorithm) and GATK (version 2.5.2) 32,33 . Pileup files were generated by SAMtools, and the allele frequency was calculated as the proportion of reads presenting the allele amongst all reads mapped to this position 34 .

HRV-A21 genome amplification. Primers targeting the complete genome of HRV-A21 were designed according to the HRV-A21 sequences obtained from deep sequencing and used to amplify genomic sequences from other HRV-A21-positive samples (see Supplementary Table S2 ). The 5′ and 3′ UTR sequences were determined via the RACE System (Invitrogen, Carlsbad, CA, USA) according to the manufacturer's protocol. Total RNA from respiratory specimens was converted to cDNA using combined random primers, oligo(dT) primers, and the SuperScript III reverse transcription system (Invitrogen, Carlsbad, CA). PCR was performed using the following conditions: 94 °C for 5 min, 40 cycles of amplification at 94 °C for 30 s, 50 °C for 45 s, and 72 °C for 90 s, with a terminal elongation step at 72 °C for 10 min. PCR products were sequenced directly and assembled manually through alignment to the reference strain (GenBank accession number KM576764). The whole genome and viral protein genes were used to construct phylogenetic trees in MEGA 5.1 using the neighbor-joining method with Kimura's two-parameter model and 1,000 bootstrap pseudo-replicates 30 . Virus isolation and neutralization assay. The HRV-A21 virus was isolated at 33 °C using a human cervical carcinoma cell line (H1-Hela; ATCC CRL-1958) from BALF samples of patient RMH001 collected on day 3 after the onset of symptoms. Cytopathic effects could be visualized. A neutralization assay was then carried out by using the reference antiserum against the prototype HRV-A21 strain (ATCC VR-1131 AS/GP) and the HRV-A21 isolate in accordance with the World Health Organization (WHO) standard procedures for poliovirus 35 . Nucleotide sequence accession numbers. The partial sequences obtained from genotyping were deposited in GenBank under the accession numbers: KR871703, KR871703, KR871714, KT229617, KT229619, KT229620, KU672654, KU859961, KU859962, KR873691 to KR873699, KR871693 to KR871699, KX384582 to KX384588, and KR871706 to KR871709. The GenBank accession numbers of the full HRV-A21 genomes obtained are KM576764 to KM576767, KR871674 to KR871677, and KR090587. The project number of the deep sequencing data is PRJNA335944.

Statistical analyses. Pearson's Chi square tests or Fisher's exact tests were used for categorical variables and Mann-Whitney U tests was used to compare continuous variables by using R version 2.15.3 (R Core Team. R: A language and environment for statistical computing. R Foundation for Statistical Computing, Vienna, Austria). P < 0.05 was considered statistically significant.

Global, regional, and national age-sex specific all-cause and cause-specific mortality for 240 causes of death, 1990-2013: a systematic analysis for the Global Burden of Disease Study

Viral pneumonia

Surveillance for emerging respiratory viruses

A novel outbreak enterovirus D68 strain associated with acute flaccid myelitis cases in the USA (2012-14): a retrospective cohort study

Virus-induced exacerbations in asthma and COPD

Fatal respiratory infections associated with rhinovirus outbreak

Role of rhinovirus load in the upper respiratory tract and severity of symptoms in lung transplant recipients

Community-Acquired Pneumonia Requiring Hospitalization among US Adults

Clinical and molecular characterization of rhinoviruses A, B, and C in adult patients with pneumonia

Prospective genotyping of human rhinoviruses in children and adults during the winter of 2009-2010

Outgrowth of the bacterial airway microbiome after rhinovirus exacerbation of chronic obstructive pulmonary disease

The co-pathogenesis of influenza viruses with bacteria in the lung

Rhinovirus genome variation during chronic upper and lower respiratory tract infections

Risk factors for severe outcomes following 2009 influenza A (H1N1) infection: a global pooled analysis

Clinical findings in 111 cases of influenza A (H7N9) virus infection

Epidemiological, demographic, and clinical characteristics of 47 cases of Middle East respiratory syndrome coronavirus disease from Saudi Arabia: a descriptive study

A novel finding for enterovirus virulence from the capsid protein VP1 of EV71 circulating in mainland China

Genome sequence of enterovirus d68 from

Recent advances from studies on the role of structural proteins in enterovirus infection

Attenuating mutations in coxsackievirus B3 map to a conformational epitope that comprises the puff region of VP2 and the knob of VP3

Introduction of a strong temperature-sensitive phenotype into enterovirus 71 by altering an amino acid of virus3D polymerase

Next-generation sequencing technologies in diagnostic virology

Clinical and epidemiological characteristics of a fatal case of avian influenza A H10N8 virus infection: a descriptive study

Infectious Diseases Society of America/American Thoracic Society consensus guidelines on the management of community-acquired pneumonia in adults

Simplification of the IDSA/ATS criteria for severe CAP using meta-analysis and observational data

Genetic clustering of all 102 human rhinovirus prototype strains: serotype 87 is close to human enterovirus 70

Simultaneous detection of fourteen respiratory viruses in clinical specimens by two multiplex reverse transcription nested-PCR assays

Sequencing and analyses of all known human rhinovirus genomes reveal structure and evolution

MEGA5: molecular evolutionary genetics analysis using maximum likelihood, evolutionary distance, and maximum parsimony methods

MEGAN analysis of metagenomic data

Fast and accurate long-read alignment with Burrows-Wheeler transform

A framework for variation discovery and genotyping using next-generation DNA sequencing data

The Sequence Alignment/Map format and SAMtools

Standard procedure for determining immunity to poliovirus using the microneutralization test. WHO/EPI/GEN/93.9

We would like to thank Caroline Tapparel