Introduction to SRA Aligned Read Format and Cloud Metadata for SARS-CoV-2

Contributors

Jon Trow

Questions

• What is the SRA Aligned Read Format

• How can I use it?

Objectives

• Understand which data and metadata types are available for SARS-CoV-2 SRA runs in the cloud.

• Understand what SRA Aligned Read Format is and what can you do with it.

• Working knowledge of SRA Metadata available on the cloud, ways to access it, and some popular use cases.

Why have SRA data in the cloud?

• SRA data archive format is over 18PB (great!)

  …but it’s a lot of data that is hard to search (not so great)

• Do you wish:

  • It was easier to find SRA data based on organismal content?
  • Searching SRA based on submitted BioProject and BioSample information was simpler?
  • You could get a sense of what could be assembled out of an SRA dataset before downloading?
  • You didn’t need to do the assembly/variant calling yourself?

What data is in scope for this training?

• Public SRA data that contains SARS-CoV-2 sequence
  • Illumina platform only
  • Stay tuned for long-read data
• How do we determine which runs contain SARS-CoV-2 data?
  • SRA Taxonomy Analysis Tool (STAT)

  • Preprint: doi:10.1101/2021.02.16.431451v1

Introduction to SRA aligned read format and associated metadata

• Incoming SRA runs are scanned for Coronaviridae content
• Contigs are created via guided assembly against SARS-CoV-2 RefSeq - fasta contigs
• Reads are mapped back to the contigs - aligned sam, fasta/q reads
• Taxonomy assessed
  • STAT
  • BLAST
• Variants are called then annotated with VIGOR3 - VCF

Using aligned read format

.pull-left[

• Coronaviridae contigs constructed from SRA data
  • Compatible with SRA Toolkit
  • Access to reads and contigs individually
  • Ability to see which reads align to each contig
  • Smaller size than standard SRA format- no quality scores ]

.pull-right[ ]

Accessing SRA metadata in the cloud

• Cloud query services- BigQuery and Athena
  • Analyze SRA metadata directly using standard SQL, allowing searches that are not possible in Entrez
  • No need to manually transform the data for querying, so you can be up and running in just a few minutes
  • Built in support for complex data fields
  • You pay for the data scanned, though some free options exist
• The underlying metadata files are available for download at no cost
  • JSON format can be transformed according to your preferences
  • These can be imported directly to Galaxy

Available SRA metadata

SRA aligned read metadata:

• User submitted metadata
  • Biological sample and sequencing library metadata supplied at the time of sequence submission
• Contig metadata
  • Contig stats including: coverage, taxonomy ID, contig length
• Taxonomic content analysis of contigs
  • Search for records based on kmer hits to tax level of interest
• BLAST results for contigs
  • Contig blast results including: hit accession, hit length, percent identity
• Variant calls
  • Variation and VIGOR3 annotation information
• Peptides
  • Details of annotated peptides, including sequence

Popular use cases- what can you do with it?

• Finding SARS-CoV-2 contigs with specific coverage and length
• Filter SARS-CoV-2 contigs based on geographic location of sample collection
• Find top SARS-CoV-2 BLAST hits for contigs from a specific geographic location
• Filter for all variations called from samples collected in the United States after 02/01/2021
• Filter for runs with the E484K mutation

SRA data in the cloud

.pull-left[

• COVID-focused SRA datasets available in GCP and AWS
  • Searchable metadata
  • SRA aligned read files
  • VCF files
  • Full SRA runs
  • Original format submitted files
• Benefits
  • Access large volume of COVID-19-related raw read data with no throttling, at no cost to you
  • Can be imported into Galaxy for use in your existing workflows ]

.pull-right[ .image-100[] .image-100[] ]

Key Points

Thank you!