Nextstrain build for novel coronavirus SARS-CoV-2
Source code
Contributors
Latest commits
Pages

Running the analysis

This section focuses on how to running the basic example build to give you a chance to practice and get a sense of how things work. The next section covers customizing and configuring your own build.

To run our analyses, we need to: 1. Ensure our sequence data and metadata is properly formatted 2. Specify which builds you want to generate using a builds.yaml file 3. Execute the workflow 4. [Hopefully you don't have to] troubleshoot

Step 1. Gather and format your data

If you haven't done this step yet, check out our data prep guide and come back when you're ready.

Step 2. Specify which builds to run

In the orientation section, we learned that - Nextstrain analyses are run using a workflow manager called Snakemake - A "build" is a bundle of input files, parameters, and commands - Each build is primarily configured by your builds.yaml file: builds.yaml and config.yaml

Let's start with defining a build in ./my_profiles/example/builds.yaml. We use the builds.yaml file to define what geographic areas of the world we want to focus on. Each block in this file will produce a separate output JSON for visualization.

The first block of the provided ./my_profiles/example/builds.yaml file looks like this:

builds:
  # Focus on King County (location) in Washington State (division) in the USA (country)
  # with a build name that will produce the following URL fragment on Nextstrain/auspice:
  # /ncov/north-america/usa/washington/king-county
  north-america_usa_washington_king-county: # name of the build; this can be anything
    subsampling_scheme: location # what subsampling method to use (see parameters.yaml)
    region: North America
    country: USA
    division: Washington
    location: King County
    # Whatever your lowest geographic area is (here, 'location' since we are doing a county in the USA)
    # list 'up' from here the geographic area that location is in.
    # Here, King County is in Washington state, is in USA, is in North America.

Looking at this example, we can see that each build has a:

  • build_name, which is used for naming output files
  • subsampling_scheme, which specifies how sequences are selected. Default schemes exist for region, country, and division. Custom schemes can be defined.
  • region, country, division, location: specify geographic attributes of the sample used for subsampling

The rest of the builds defined in this file serve as examples for division-, country- or region-focused analyses. To adapt this for your own analyses:

  1. copy my_profiles/example to my_profiles/<my-new-name>
  2. open and modify the builds.yaml file in this directory to include your geographic area(s) of interest and remove any builds that are not relevant to your work
  3. open and modify the config.yaml file in this directory such that it references:
    • the path to your new custom builds.yaml instead of the example builds file
    • the path to your own sequences and metadata instead of the example data

Step 3: Run the workflow

To actually execute the workflow, run:

ncov$ snakemake --profile my_profiles/example -p

--profile tells snakemake where to find your builds.yaml and config.yaml files. -p tells snakemake to print each command it runs to help you understand what it's doing.

If you'd like to run a dryrun, try running with the -np flag, which will execute a dryrun. This prints out each command, but doesn't execute it.

Note that the example profile runs the workflow with at most two cores at once, as defined by the cores parameter in my_profiles/example/config.yaml. Snakemake requires you to specify how many cores to use at once. To define the number of cores to use from the command line, run Snakemake as follows.

ncov$ snakemake --cores 1 --profile my_profiles/example -p

Step 4: Troubleshoot common issues

If you have a question which is not addressed here, please don't hestitate to ask for help

My country / division does not show up on the map

This is most often a result of the country / division not being present in the file defining the latitude & longitude of each deme. Adding it to that file (and rerunning the Snakemake rules downstream of this) should fix this.

My trait (e.g. division) is grey instead of colored

We generate the colors from the colors rule in the Snakefile, which uses the ordering TSV to generate these. See 'customizing your analysis' for more info.

A note about locations and colors: Unless you want to specifically override the colors generated, it's usually easier to add information to the default ncov files, so that you can benefit from all the information already in those files.

My genomes aren't included in the analysis

There are a few steps where sequences can be removed:

  • During the filter step:
    • Samples which are included in the exclude file are removed
    • Samples which fail the current filtering criteria, as defined in the parameters.yaml file, are removed. You can modify the snakefile as desired, but currently these are:
      • Minimum sequence length of 25kb
      • No ambiguity in (sample collection) date
    • Samples may be randomly removed during subsampling; see 'customizing your analysis' for more info.
    • During the refine step, where samples that deviate more than 4 interquartile ranges from the root-to-tip vs time are removed

Error: Where there's SAMPLING_TRAIT we should always have EXPOSURE_TRAIT

This comes from an incomplete metadata file. If you define (e.g.) country for a sample then you must also define country_exposure for that sample. If there is no (known) travel history, then you can set the same values for each.

Sequencing and alignment errors

Genome sequencing, bioinformatic processing of the raw data, and alignment of the sequences are all steps were errors can slip in. Such errors can distort the phylogenetic analysis. To avoid sequences with known problems to mess up the analysis, we keep a list of problematic sequences in config/exclude.txt and filter them out. To facilitate spotting such problematic sequences, we added an additional quality control step that produces the files

  • results/sequence-diagnostics.tsv
  • results/flagged-sequences.tsv
  • results/to-exclude.txt

These files are the output of scripts/diagnostics.py and are produced by rule diagnostic. The first file contains statistics for every sequence in the aligment, sorted by divergence worst highest to lowest. The second file contains only those sequences with diagnostics exceeding thresholds each with their specific reason for flagging -- these are sorted by submission date (newest to oldest). The third file contains only the names of the flagged sequences and mirrors the format of config/exclude.txt. These names could be added to config/exclude.txt for permanent exclusion. Note, however, that some sequences might look problematic due to alignment issues rather than intrinsic problems with the sequence. The flagged sequences will be excluded from the current run.

To only run the sequence diagnostic, you can specify any of the three above files as target or run: bash snakemake --profile my_profiles/<name> diagnostic

In addition, we provide rules to re-examine the sequences in config/exclude.txt. By running bash snakemake --profile my_profiles/<name> diagnose_excluded the pipeline will produce

  • results/excluded-sequence-diagnostics.tsv
  • results/excluded-flagged-sequences.tsv
  • results/check-exclusion.txt

These files are meant to facilitate checking whether sequences in config/exclude.txt are excluded for valid reasons.

Previous Section: Orientation: which files should I touch?

Next Section: Orientation: Customizing your analysis