DNAStack COVID19 SRA Data

The Sequence Read Archive (SRA)  is the primary archive of high-throughput sequencing data, hosted by the National Institutes of Health (NIH). The SRA represents the largest publicly available repository of SARS-CoV-2 sequencing data. This dataset was created by DNAstack using SARS-CoV-2 sequencing data sourced from the SRA. Where possible, raw sequence data were processed by DNAstack through a unified bioinformatics pipeline to produce genome assemblies and variant calls. The use of a standardized workflow to produce this harmonized dataset allows public data generated using different methodologies to be combined and compared for a more powerful global analysis of available SARS-CoV-2 data, allowing researchers rapid access to aggregated downstream results for accelerated insight generation. Methodology: Reads from the SRA were extracted in FASTQ format, then entered into a different pipeline depending on the sequencing technology used to create the reads: the ARTIC protocol  for Oxford Nanopore-derived reads; the SIGNAL pipeline  for paired-end Illumina reads; and the CoSA pipeline  (using DeepVariant  for variant calling) for PacBio reads. Briefly, reads were primer-trimmed and aligned to the SARS-CoV-2 reference genome, following which contiguous regions were assembled and variant sites were called. Pangolin  was then used to assign viral lineage based on the assembled genome.