We've prepared an example dataset in the
data directory. If you'd like to move ahead with this tutorial with this example dataset, you can skip this section.
If you'd like to use your own data, read on.
To use Nextstrain to analyze your own data, you'll need to prepare two files:
fastafile with viral genomic sequences
tsvfile with metadata describing each sequence
We've created an example dataset in the
data directory. This consists of a compressed FASTA file with viral genomes sourced from Genbank (
example_sequences.fasta.gz) and a corresponding TSV with metadata describing these sequences (
The first 2 lines in
data/sequences.fasta look like this:
The first line is the
name of the sequence.
Lines with names in FASTA files always start with the
> character (this is not part of the name), and may not contain spaces or
Note that "strain" here carries no biological or functional significance and should largely be thought of as synonymous with "sample."
The sequence itself is a consensus genome.
By default, sequences less than 27,000 bases in length or with more than 3,000
N (unknown) bases are omitted from the analysis.
For a basic QC and preliminary analysis of your sequence data, you can use clades.nextstrain.org.
This tool will check your sequences for excess divergence, clustered differences from the reference, and missing or ambiguous data. In addition, it will assign nextstrain clades and call mutations relative to the reference.
Nextstrain accommodates many kinds of metadata, so long as it is in a
TSV is a text file, where each row (line) represents a sample and each column (separated by tabs) represents a field.
If you're unfamiliar with TSV files, don't fret; it's straightforward to export these directly from Excel, which we'll cover shortly.
Here's an example of the first few columns of the metadata for a single strain, including the header row. (Spacing between columns here is adjusted for clarity, and only the first 6 of 23 columns are shown).
strain virus gisaid_epi_isl genbank_accession date region ... NewZealand/01/2020 ncov EPI_ISL_413490 ? 2020-02-27 Oceania ...
In total there are 23 columns of metadata for each genome; see the last section of this page for an in-depth walkthrough.
A valid metadata file must include the following fields:
||Sample name / ID||Each header in the fasta file must exactly match a
||Date of sampling||
||Pathogen name||Needs to be consistent|
||Global region of sampling|
Please be aware that our current pipeline will filter out any genomes with an unknown date - you can change this in your own pipeline.
Missing data is to be expected for certain fields.
In general, missing data is represented by an empty string or a question mark character.
There is one important difference: if a discrete trait reconstruction (e.g. via
augur traits) is to be run on this column, then a value of
? will be inferred, whereas the empty string will be treated as missing data in the output. See below for how to represent uncertainty in sample collection date.
You can also create a TSV file in Excel. However, due to issues with auto-formatting of certain fields in Excel (like dates), we don't recommend this as a first option. If you do edit a file in Excel, open it afterwards in a text-editor to check it looks as it should!
File > Save as
Text (Tab delimited) (*.txt)and enter a filename ending in
Making inferences about a sample's origin is strongly dependent on the makeup of your dataset: the model can't infer a transmission from an origin it doesn't have any (or enough) data from.
To address this, we strongly recommend adding contextual background sequences to your dataset. To make this easier, we provide a continually-updated dataset, pre-formatted for Nextstrain, through GISAID. To download this dataset:
metadata_2021-04-08_08-30.tsv.gz. Uncompress this file and save as
sequences_2021-04-08_08-30.fasta.gz. Keep this file compressed and save as
GISAID maintains multiple tiers of access and many users of GISAID will initially lack access to the "FASTA" and "metadata" under "Genomic epidemiology" (the entries will just be absent). If you find that these files are missing for you, you'll need to email GISAID at hCoVfirstname.lastname@example.org to request access to the specific "FASTA" and "metadata" files under "Genomic epidemiology". Somewhat confusingly, there are separate "FASTA" and "metadata" entries listed under the heading "Download packages". These files are a slightly different format and are not directly compatible with the Nextstrain ncov pipeline. Please feel free to let us know at discussion.nextstrain.org if you're having difficulties accessing these files.
The Nextstrain team uses this pipeline to include the latest sequences and metadata from GISAID in our builds: nextstrain/ncov-ingest.
We've outlined several methods for subsampling, including builds with a focus area + genetically similar contextual sequences, in the next section.
This needs to match the name of a sequence in the FASTA file exactly and must not contain characters such as spaces, or
In our example we have a strain called "NewZealand/01/2020" so there should be a sequence in the FASTA file for ">NewZealand/01/2020" (sequence names in FASTA files always start with the
> character, but this is not part of the name).
Note that "strain" here carries no biological or functional significance and should be thought of as synonymous with sample.
Name of the pathogen.
If this genome is shared via GISAID then please include the EPI ISL here. In our example this is "EPI_ISL_413490".
If this genome is shared via GenBank then please include the accession number here. In our example this is "?" indicating that it hasn't (yet) been deposited in GenBank. (See above for more information on how to encode missing data.)
date (really important!)
This describes the sample collection data (not sequencing date!) and must be formated according as
Our example was collected on Feb 27, 2020 and is therefore represented as "2020-02-27".
You can specify unknown dates or month by replacing the respected values by
2011-XX-XX) and completely unknown dates can be shown with
20XX-XX-XX (which does not restrict the sequence to being in the 21st century - they could be earlier).
Please be aware that our current pipeline will filter out any genomes with an unknown date, however you can change this for your pipeline!
See this guide to formatting dates in Excel
The region the sample was collected in -- for our example this is "Oceania". Please use either "Africa", "Asia", "Europe", "North America", "Oceania" or "South America". If you sequence a genome from Antartica, please get in touch!
The country the sample was collected in. Our example, "NewZealand/01/2020", was collected in ....... New Zealand.
You can run
tail +2 data/metadata.tsv | cut -f 7 | sort | uniq to see all the countries currently present in the metadata.
As of April 10, there were 64! 🌎
Division currently doesn't have a precise definition and we use it differently for different regions. For instance for samples in the USA, division is the state in which the sample was collected here. For other countries, it might be a county, region, or other administrative sub-division. To see the divisions which are currently set for your country, you can run the following command (replace "New Zealand" with your country):
tail +2 data/metadata.tsv | cut -f 7,8 | grep "^New Zealand" | cut -f 2 | sort | uniq
division, but for a smaller geographic resolution. This data is often unavailable, and missing data here is typically represented by an empty field or the same value as
division is used.
In our example the division is "Auckland", which conveniently (or confusingly) is both a province of New Zealand and a city.
If the sample has a known travel history and infection is thought to have occured in this location, then represent this here. In our example, which represents New Zealand's first known case, the patient had recently arrived from Iran, thus the value here is "Asia". Specifying these travel histories helps inform the model we use to reconstruct the geographical movements of the virus.
If there is no travel history then set this to be the same value as
region_exposure but for
In our example, given the patient's travel history, this is set to "Iran".
region_exposure but for
division. If we don't know the exposure division, we may specify the value for
country_exposure here as well.
Unused. Typically the value "genome" is set here.
Genome length (numeric value).
Host from which the sample was collected. Currently we have multiple values in the dataset, including "Human", "Canine", "Manis javanica" and "Rhinolophus affinis".
Numeric age of the patient from whom the sample was collected. We round to an integer value. This will show up in auspice when clicking on the tip in the tree which brings up an info box.
Sex of the patient from whom the sample was collected. This will show up in auspice when clicking on the tip in the tree which brings up an info box.
Please see GISAID for more information.
Please see GISAID for more information.
Author of the genome sequence, or the paper which announced this genome. Typically written as "LastName et al". In our example, this is "Storey et al". This will show up in auspice when clicking on the tip in the tree which brings up an info box.
The URL, if available, pointing to the genome data. For most SARS-CoV-2 data this is https://www.gisaid.org.
The URL, if available, of the publication announcing these genomes.
Date the genome was submitted to a public database (most often GISAID).
YYYY-MM-DD format (see
date for more information on this formatting).