To allow public sharing of SARS-CoV-2 sequencing data, care must be made to
filter out any background human sequencing data. To combat this, a set of
workflows was designed to filter out human aligned reads providing SARS-CoV-2
sequences alone. To test this, whole genome data was downloaded from the
Genome in a Bottle project for sample NA12878 (see Data Downloads section)
along with viral genome sequencing data for sample NT04 from the Short Read
Archive. Selecting 10,000 reads from the NA12878 data and randomly inserting
them into the NT04, a final set of paired-end FASTQ files containing human
spiked-in SARS-CoV-2 were created.
The data used in the generated reads can be found here:
ftp://ftp-trace.ncbi.nlm.nih.gov/ReferenceSamples/giab/data/NA12878/NIST_NA12878_HG001_HiSeq_300x/RMNISTHS_30xdownsample.bam
https://trace.ncbi.nlm.nih.gov/Traces/sra/?study=SRP253798
Pre-generated data is provided in the data directory and includes SARS-CoV-2 reads from host NT04 with NA12878 human host reads. Aligment were filtered from the NA12878 BAM file for mapping quality > 30. Aligments were name sorted using samtools sort and limited to the first 5000 read pairs. The alignments were converted to FASTQ files using samtools fastq and added to the SARS-CoV-2 reads using the spikein_reads.py script.
MIT