33 Zika virus genomes sequenced from patient and pooled mosquito samples


33 Zika virus genomes sequenced from patient and pooled mosquito samples collected in Rio de Janeiro, Tegucigalpa, Santo Domingo, and Florida, March-September 2016

The Broad Viral Genomics Group and MIT, together with partners at Oswaldo Cruz Foundation (Fiocruz) in Brazil; National Autonomous University of Honduras, and Florida Gulf Coast University (FGCU), are releasing 33 near-complete Zika virus (ZIKV) genomes sequenced from patient samples and mosquito pools collected in Brazil, Honduras, the Dominican Republic, and Florida between March and September 2016.

We are releasing the assemblies as consensus genomes and are in the process of depositing assemblies and read data, as well as sample metadata (generously shared by partners and including: place and date of collection, age, sex, and date of symptom onset), on NCBI GenBank.

The 33 ZIKV consensus sequences can be downloaded here:
ZIKV_BROAD_2016-10.fasta.txt (342.5 KB; this is a fasta file ending in ‘.txt’ because virological only allows certain file extensions)

Of the 33 samples, 9 are from Rio de Janeiro, Brazil, 7 are from Tegucigalpa, Honduras, 8 are from Santo Domingo, Dominican Republic, and 9 are from Florida, United States. 29 are from Zika-infected patients (10 urine, 7 plasma, 12 serum) and 4 are from ZIKV+ mosquito pools caught in Florida. All 33 genomes have >70% of the Zika genome and 22 have >95% of the Zika genome covered.

The samples were not cultured or passaged prior to sequencing. We experimented with different unbiased library preparation methods, and/or used amplicon-based sequencing, followed by Illumina sequencing on MiSeq and HiSeq platforms at the Broad Institute. Genomes were assembled using the Sabeti lab’s viral-ngs pipeline, which creates a de novo assembly from cleaned reads and performs additional refinement in a reference-assisted scaffolding process.

We assembled 17 Zika genomes after enriching libraries for Zika content by the Sabeti lab’s hybrid capture protocol. 13 have median coverage >25x and 4 have median coverage >250x. The other 16 were sequenced with amplicon-based sequencing using a protocol developed by the ZiBRA project and modified by the Andersen lab. We are still optimizing our pre-sequencing sample protocols, and will document our full process in a future update.

For more information about this project and data release, contact dpark@broadinstitute.org.

Please feel free to download, use, and share this data. Our partners and we are currently in the process of preparing publications describing our wet lab and genome analysis methods, and describing viral evolution and transmission patterns. We will post progress regarding data improvements, publications, and terms of use on this forum. If you intend to use these sequences for publication prior to the release of our papers, please contact us directly. If you are interested in joining our collaboration—or if you have any other questions—please also contact us directly.

The work was and continues to be part of a productive and growing collaboration across nations. Those currently involved include the following investigators and their teams:

Pardis Sabeti, Broad Institute and Harvard University, USA
Irene Bosch and Lee Gehrke, MIT, USA
Thiago Moreno L. Souza, Patrícia T. Bozza, Dr. Gonzalo Belo, and Wim Degrave, Oswaldo Cruz Foundation, Brazil
Fernando Bozza, Oswaldo Cruz Foundation and D’or Institute, Brazil
Fabiano L. Thompson, Federal University of Rio de Janeiro, Brazil
Kristian Andersen, The Scripps Research Institute, USA
Salim Mattar, Universidad de Córdoba, Colombia
Ivette Lorenzana, Universidad Nacional de Honduras, Honduras
Sharon Isern and Scott F. Michael, Florida Gulf Coast University, USA
Daniel Olson and Edwin J. Asturias, Colorado School of Public Health, USA

ZIKV_BROAD_2016-10_ML-TREE.pdf (332.7 KB)


Cool! So, for the Florida sequences, are these from the same cases as those reported by Andersen et al. (i.e., the link Nathan posted yesterday)? Or unique cases?


The Florida genomes are from the same samples as ones reported by the Andersen lab. The sequences are slightly different. To prevent confusion, here is a mapping between the sample IDs we use (Broad ID) and ones used by the Andersen lab (Scripps ID):

Broad ID    Scripps ID
USA_2016_FL-010_URI    ZF10
USA_2016_FL-030_URI    ZF30
USA_2016_FL-032_URI    ZF32
USA_2016_FL-038_URI    ZF38
USA_2016_FL-039_URI    ZF39
USA_2016_FL-01_MOS    ZM1
USA_2016_FL-02_MOS    ZM2
USA_2016_FL-03_MOS    ZM3
USA_2016_FL-04_MOS    ZM4


Thanks so much for making these available. In collaboration with @richard.neher, I’ve been maintaining updated Zika sequences on nextstrain.org/zika/. I’ve added these genomes. You can see the Broad data by selecting “Broad Viral Genomics Group” under “Consortium”.

Some interesting clustering here with Honduras and DR samples.

For the moment, I’ve set dates as 2016-XX-XX, so that we estimate date of sampling in 2016 via molecular clock in TreeTime. Would it be possible to get collection dates for these sequences? It would definitely improve the time tree analysis.

Also, please let me know if you’d like any attribution changed on the website.


Small note that should have mentioned: I’ve left out all the Broad USA samples from the nextstrain tree as all these samples already existed in the tree via Scripps as noted above: http://virological.org/t/33-zika-virus-genomes-sequenced-from-patient-and-pooled-mosquito-samples/372/3