Global Population Structure and Genomic Correlates of Severe Disease


Despite the threat of Dengue virus to public health and its importance to research on viral hemorrhagic fevers our understanding of the virulence, pathogenesis, and mechanisms of re-emergence of the dengue virus is limited.  As a consequence, presently no vaccine or anti-viral therapy exists for this flavivirus, and the genetic underpinnings of disease outcomes are unknown.  Studies of nucleotide divergence among he different serotypes has largely been limited to a single gene.  This lack of basic information of viral diversity severely limits vaccine and anti-viral therapy development efforts.  In addition, the presence of immune pathology in secondary DEN infections indicates that previous non-sterile immune response to one serotype can exacerbate the immune response to the secondary infections caused by another serotype.  This host-viral interaction also represents a considerable challenge for vaccine development. 

Through this project we intend to sequence >3500 dengue genomes of distinct geographic origin and disease to build the genomic infrastructure needed to study and combat this virus.  We will define the population structure of the virus at multiple scales (i.e. within host, local, regional, and continental) which will enable determination of the impact of introduced strains versus indigenous evolution on disease outcomes.  Additionally, this project aims to understand the genomic correlates of disease severity and will provide the first map of genomic distributions with reference to DF, DHF, and DSS. 


Principal External Collaborators

Irene Bosch - University of Massachusetts Medical School, Center for Infectious Disease and Vaccine Research, Worcester, MA

Philippe Buchy - Virology Unit, Institut Pasteur in Cambodia, Ohnom Penh, Cambodia

Eva Harris - Division of Infectious Diseases, School of Public Health, University of California, Berkeley, CA

Jorge Munoz - Centers for Disease Control and Prevention, Division of Vector-Borne Infectious Disease, San Juan, PR

Cameron Simmons - Oxford University Clinical Research Unit, Hospital for Tropical Diseases, Ho Chi Minh City, VietNam

Priscilla Yang - Dept. of Microbiology, Harvard Medical School, Boston, MA


Funding for this sequencing project was provided by the National Institute of Allergy and Infectious Diseases.

Cohort Descriptions

Cohort / Sample Set Code: AR

Primary GRID Collaborator: Irene Bosch, Ph.D.

These samples consist of four serotypes collected in the Center/North region of Venezuela over a five-year period. Transmission rates are very high in this area due to the presence of large water reservoirs and a high mosquito population. It is a known hyper-endemic area of transmission of Dengue with co-circulation of the four serotypes. The cohort will be used to evaluate the population genetic structure of dengue in a region with high transmission rates. Viruses isolated from this region of Venezuela will be contrasted and compared to viruses isolated in the capital, as well as regions geographically isolated by the Andes mountains.

Cohort / Sample Set Code: CA

Primary GRID Collaborator: Irene Bosch, Ph.D.

Caracas, Venezuela, 6-year clinical prospective study 2001-2007
This cohort consists of predominantly DENV-3 isolates collected in the Capital region of Venezuela.  These samples span several epidemic periods, 2001-2002 and 2007, where DENV transmission rates are very high.  Patients in this fever cohort were intensively evaluated, providing extensive clinical information for patient phenotyping; about 100 acute patients were evaluated daily. The spectrum of disease in this cohort is limited to dengue fever with severe manifestations of disease not predominant until 2007. During the 2007 epidemic, an increase in severity was observed in patients infected with DENV-3.  Dengue Shock Syndrome has not been observed in this cohort.

Other info

~300 samples; 100 with acute high res info
It is a study of evolution over time, as the location is constrained to the Caracas metropolitan area. The total population of Caracas is approximately 5 million, with very vast areas of dense tropical vegetation and poor housing projects with possibly high transmission rates. Around 70% of the population is under 18 years old, a typical demographic pyramid shape, and this large group of immunologically naive individuals represents a potential factor contributing to the high transmission rates.

Cohort / Sample Set Code: ME

Primary GRID Collaborator: Irene Bosch, Ph.D.

DENV isolates in this sample set represent the 'A-group' collected in the western part of Venezuela. These samples provide an opportunity to study genetic diversity in a DENV population that has likely been isolated from strains circulating in other regions of the country by the Andes mountains, but was not isolated from strains circulating in Colombia.

Other info:

The serotypes were collected over a period of multiple years, and the total number of samples is close to 40.

Cohort / Sample Set Code: KP

Primary GRID Collaborator: Irene Bosch, Ph.D.

DENV isolates in this cohort were collected in the northern part of Thailand (Kamphaeng Phet). This site has been studied for a long time by the USA Army Institute based in Thailand and is a designated site for a future dengue vaccination trial. Isolates from this area represent DENV-1,2,3 infections that occurred over a one-year period and are segregated among school districts.  Mosquito biting amongst school classmates represents a potential clustering of infections whereby constraints on viral evolution can be evaluated in the context of clustered transmission events.  Sample size is 70.

Cohort / Sample Set Code: Thai, Colombia, UNC, UMMS

Primary GRID Collaborator: Irene Bosch, Ph.D.

University of Massachusetts Medical School Dengue Collection
This sample set includes laboratory archival samples that span several generations of laboratory evolution, geographical sites, and times of collection. These isolates represent a large archival sample set and as such are available to researchers for future studies. Genomes in this collection are from various parts of the world, including Sri Lanka, Peru, Thailand (DENV-4 isolates included), and China.

Cohort / Sample Set Code: BR

Primary GRID Collaborator: Irene Bosch, Ph.D.

This sample set consists of DENV-1, DENV-2, and DENV-3 isolates collected over several years in various regions of Brazil where there is no reported incidence of infections with DENV-4. Two samples of each year, of each serotype, dating from 2000-2008.

Cohort / Sample Set Code: YU-MEX and MEX

Primary GRID Collaborator: Irene Bosch, Ph.D.

YU-MEX: This sample set consists of DENV isolates (1, 2, and 4) representing  three of the serotypes collected in the Yucatan. The collection years are from 2005 to 2008. The total number of isolates is 206.
MEX: Seroptypes 1-4 from various geographical regions of Mexico, representing the years between 2002-2007. The number of samples collected is 183.

Cohort / Sample Set Code: RP-BR

Primary GRID Collaborator: Irene Bosch, Ph.D.


This sample set consists of DENV-1, 2 and 3 isolates collected in Brazil from 2006-2008 during a period of high transmission in ONE region of Brazil. These samples were collected with high-resolution geographical positioning information and are suitable to evaluate disease clustering in this region. The number of samples is 200.

Cohort / Sample Set Code: SAN-CO

Primary GRID Collaborator: Irene Bosch, Ph.D.

100 samples from Santander-Colombia, from 1998-2008. Serotypes 1,2,3, and 4. This is a cohort of samples to study evolution of Dengue through time in this region of Colombia. Colombia has three mountain ranges that divide the country into three regions along a North-South axis. Santander belongs to the Andes region and is an �isolated area� in the main continent; it is surrounded by mountains on the South, East and West.   

Cohort / Sample Set Code: DENFRAME

Primary GRID Collaborator: Philippe Buchy, Ph.D.


The DENFRAME consortium is a multidisciplinary research group (13 institutions) funded by the European Union, which is dedicated to the development of new diagnostic tools and therapeutic approaches for Dengue disease. The Consortium integrates clinicians, epidemiologists, virologists and researchers from other disciplines.
This study is a prospective epidemiological study of Dengue fever patients and their household members. In Cambodia, samples were collected in the paediatric unit of a referral hospital in Kampong Cham province and sent to Institut Pasteur in Phnom Penh to be tested for Dengue infection by a combination of serological, virological and molecular techniques.
The aims of the study were:

  • To standardize the procedures for current diagnostic assays;
  • To implement inter-laboratory control procedures;
  • To establish a bank of clinical samples to validate rapid and sensitive antigen detection (NS1 for this study) and/or serological tests (using ligand binding molecules and chemiluminescent biosensor techniques) to be developed as part of the DENFRAME project;
  • To estimate the proportion of non or pauci-symptomatic Dengue infections among household members of dengue index case;
  • To generate a comprehensive epidemiological dataset to identify non-genetic confounding factors & to prepare a study on the association of human genetic polymorphisms in candidate genes with Dengue susceptibility and severity.


Cohort / Sample Set Code: ICMMAPL

Primary GRID Collaborator: Philippe Buchy, Ph.D.

The identification of cellular and molecular markers associated with plasma leakage and Dengue shock syndrome study (ICMMAPL) is a complementary research project integrated in the work-package 5 of DENFRAME project, focusing on dendritic cell (DC) interactions with the Dengue virus. The main objective is aimed at identifying pathological mechanisms supporting the endothelial dysfunction leading to plasma leakage, and finding new prognostic markers detectable earlier in the course of infection that may predict the evolution of Dengue-infected patients towards Dengue shock syndrome.

Samples were collected at the referral hospital in Kampong Cham Province. Acute and convalescence samples were collected at admission and at discharge from hospital. From June to November 2007, 71 patients with Dengue infection of different degrees of severity, and 20 patients without Dengue or other flaviviruses infection (NDC) were identified at Institut Pasteur of Cambodia in Phnom Penh.

Cohort / Sample Set Code: Dengue Surveillance

Primary GRID Collaborator: Philippe Buchy, Ph.D.

"Dengue Surveillance" is the Cambodian National Dengue Surveillance System conducted by the National Center for Malariology, Parasitology and Entomology of Cambodia with laboratory diagnosis performed at the Institut Pasteur of Cambodia since 1999. Selected samples are collected from Dengue suspect patients hospitalized in 5 sentinel sites throughout the country (Kampong Cham, Siem Reap, Battambang, Takeo and Phnom Penh) and sent back to Institut Pasteur for diagnosis and storage. The aim of this surveillance system is to monitor the circulation of the four Dengue serotypes and understand the epidemiological pattern of the infection throughout Cambodia.

Cohort / Sample Set Code: PDVI-IPC

Primary GRID Collaborator: Philippe Buchy, Ph.D.

Since May 2006, active community-based surveillance of fever illnesses and dengue fever was conducted in 25 conveniently selected villages (semi-urban and rural areas) in Kampong Cham province, Cambodia following a cohort of ~8,000 individuals > aged 19 years or younger each year. The surveillance periods span each year from May to December which represent the dengue season. Fever events were identified by home visits and mothers' reports and further investigated for acute DENV-infection using molecular and serological methods. Each fever episode was documented with demographic information and illness-related symptoms using a standardized case report form. Information related to type of care sought for the fever episode were also collected during the convalescent phase (i.e. 14 - 21 days of disease onset).

The aims of the study include the estimation of disease incidences and description of the dynamics of transmission and serotypes circulation. The site could be utilized as potential field site for a vaccine trial.

Cohort / Sample Set Code: PDVI NI

Primary GRID Collaborator: Eva Harris, Ph.D.

The Pediatric Dengue Cohort Study (PDCS) follows 3,800 children in District II of Managua, Nicaragua and has been ongoing since August of 2004. Healthy blood samples are collected every July, and during the year, children are followed for all medical consults (including all febrile illnesses) at the Health Center Socrates Flores Vivas (HCSFV). Suspected dengue cases are followed clinically at the HCSFV and at the study hospital (Hospital Infantil Manuel de Jesus Rivera) when necessary, and acute and convalescent samples undergo serological, virological, and molecular biological work-up for dengue diagnosis and isolation of dengue virus at the National Virology Laboratory in the Centro Nacional de Diagnostico y Referencia at the Ministry of Health. Information and communication technologies (ICT) have been integrated into all levels of the study to ensure the highest level of sample and data quality. The aims of the study are to investigate the natural history of dengue transmission (clinical manifestations, viral sequence), collect biological samples for immunological studies and vaccine safety research, and prepare the site for potential clinical trials of tetravalent dengue vaccines.

Cohort / Sample Set Code: Hospital

Primary GRID Collaborator: Eva Harris, Ph.D.

This Nicaraguan hospital-based study has been ongoing since 1998, and in its current form since 2005. Children, 6 months to 14 years of age, suspected of Dengue who present to the Nicaraguan National Pediatric Reference Hospital, the Hospital Infantil Manuel de Jesus Rivera, are invited to participate in the study, which follows children (as inpatients and outpatients) daily during the acute phase until discharge and again at convalescence for clinical symptoms and signs, fluid management, and collection of biological samples (serum, plasma, PBMCs, DNA). These samples are used for clinical management (complete blood count, blood chemistry) and dengue serology, virology, and molecular biology for both diagnostic and research purposes. The aims of the study are to improve Dengue case management and to better understand the pathogenesis of severe dengue using tools such as full-length genome sequencing, microarray, studies of antibody repertoire, and the like. As in the PDCS, ICT tools and clinical mononitoring have been widely implemented to ensure quality control and high-quality clinical data and biological samples.

Cohort / Sample Set Code: Routine

Primary GRID Collaborator: Eva Harris, Ph.D.

Routine samples are selected from the ongoing National Dengue Surveillance System at the Nicaraguan Ministry of Health. This surveillance system monitors ongoing Dengue outbreaks throughout the country, and selected samples are worked up by the National Virology Laboratory using serological, molecular biological, and virological assays to detect Dengue virus infection. Some of the confirmed dengue cases that are positive for virus isolation and were collected from 2005 onwards are sequenced in an effort to monitor Dengue virus diversity throughout the country.

Cohort / Sample Set Code: Archive

Primary GRID Collaborator: Eva Harris, Ph.D.

Dengue viruses of all four serotypes that have been isolated since 1998 form part of the Nicaraguan dengue archive collection.  Selected samples from this collected are being sequenced to determine the past history of Dengue virus circulation and diversity in the country (focusing mostly in the capital city of Managua).

Cohort / Sample Set Code: Cluster

Primary GRID Collaborator: Eva Harris, Ph.D.

The Index Cluster Study was conducted in Managua, Nicaragua, in December 2006-January 2007 and September 2007-November 2007. Suspected Dengue cases from the Pediatric Dengue Cohort Study were chosen (4 negative and 18 positive index cases), and 20-25 neighbors living in a 25-meter radius were recruited into the study within 48-72 hours after symptom onset of the index case. Samples were collected from the 495 contacts on days 1, 3, and 15 and tested for Dengue virus infection by serology on paired samples. Acute samples of individuals that were positive for a current yet asymptomatic  Dengue virus infection (12) were processed by RT-PCR and virus isolation, and those positive for virus isolation are being sequenced to compare viruses from asymptomatic versus symptomatic Dengue virus infections.

Cohort / Sample Set Code: C Cases

Primary GRID Collaborator: Eva Harris, Ph.D.

The Pediatric Dengue Cohort Study focuses on possible Dengue cases, which are defined as those that meet the WHO dengue case definition (A cases) or undifferentiated fever cases without a specific clinical diagnosis (B cases). C cases are febrile cases with a specific clinical diagnosis, yet testing of a certain percent of C cases has resulted in several Dengue-positive individuals. Viruses isolated from these C cases are being sequenced to determine if there are any differences as compared to Dengue viruses that cause classic Dengue fever or undifferentiated fever.

Cohort / Sample Set Code: Puerto Rico

Primary GRID Collaborator: Jorge Munoz, Ph.D.

The Dengue Branch (DB) in San Juan, Puerto Rico, is part of the Division of Vector-Borne Infectious Diseases (DVBID) of the Centers for Disease Control and Prevention (CDC), and a collaborating center for WHO and PAHO. The DB develops and maintains effective surveillance for Dengue viruses and their mosquito vectors, conducting field and laboratory research and epidemic investigations. Through these activities, the DB defines disease etiology, ecology, and pathogenesis, and provides diagnostic and reference services and consultation to Puerto Rico and other state laboratories. These efforts have rendered a collection encompassing 20 years of Dengue virus isolates associated with detailed geographical, temporal, epidemiological and clinical observations.

Cohort / Sample Set Code: CDC Reference Collection

Primary GRID Collaborator: Jorge Munoz, Ph.D.

Virus isolates from serum specimens collected by the CDC-Dengue Branch through surveillance and reference activities aim to characterize transmission of Dengue virus in Puerto Rico and the larger surrounding area. The virus was isolated in mosquito C636 cells or other cells as indicated and grown in laboratory conditions at a low passage. This characterization and the sequence analysis of the viral RNA are useful in documenting the introduction, maintenance and spread of Dengue virus strains.

Cohort / Sample Set Code: DENCO

Primary GRID Collaborator: Cameron Simmons, Ph.D.

DENCO is a multicentre study, supported by EU funds and coordinated by WHO/TDR, that has documented in a standard fashion the clinical features of >2000 Dengue patients in 8 countries (4 in South East Asia, 4 in Central/South America). The aim of the study is to create an evidence base to validate the existing WHO criteria for Dengue disease classification.

Cohort / Sample Set Code: FG

Primary GRID Collaborator: Cameron Simmons, Ph.D.

The FG study is a multicentre descriptive study of early Dengue in children in 4 hospitals in southern Viet Nam. The aim of the study is to identify clinical, host genetic, viral, and immunological risk factors for Dengue shock syndrome in children enrolled within 72 hours of fever onset.

Cohort / Sample Set Code: FR

Primary GRID Collaborator: Cameron Simmons, Ph.D.

The FR study is a multicentre descriptive study of early Dengue in adults in 4 hospitals in southern Viet Nam. The aim of the study is to identify clinical, host genetic, viral, and immunological risk factors for Dengue shock syndrome in adult patients enrolled within 72 hours of fever onset.

Cohort / Sample Set Code: MD

Primary GRID Collaborator: Cameron Simmons, Ph.D.

The MD study is a descriptive study of in-patient children with Dengue at the Hospital For Tropical Diseases, HCMC, Viet Nam. The aim of this study, which has been running for 7 years, is to better understand the pathophysiology of capillary leak in children.

Cohort / Sample Set Code: sm-DVEI

Primary GRID Collaborator: Priscilla Yang, Ph.D.

This is an experimental study that looks at how GNF2, a hydrophobic ligand, can decrease viral loads in a cell.  Genomes sequenced as part of this sample set demonstrated the ability to break through the inhibition of GNF2 and lead to 10-fold or greater increases in viral loads.  Identifying the mutations in the viral genome that confer resistance to GNF2 inhibition likely will provide much needed insight into their respective binding site(s) and mechanism(s) of action. In addition, the identification of resistance mutations will have important implications for the development of Dengue entry inhibitors as potential therapeutics.

Cohort / Sample Set Code: Mouse Adapted

Primary GRID Collaborator: Eva Harris, Ph.D.

Human clinical isolates have been serially passaged in mosquito C6/36 cells and inoculated into mice to select for virus mutants that are adapted to grow in mice. Virus in mouse sera is collected, passaged in C6/36 cells and re-inoculated into mice repeatedly. The process is continued until a phenotype of increased pathogenesis/lethality in the mice is achieved.