First we download the reads they mention in the paper using Fasterq-dump
fasterq-dump ERR1091914
This will download one of the samples in the paper that is listed as being contaminated with SARs-CoV-1.
This sample will be downloaded as a fastq file. This raw sequencing still has illumina primer sequences on them and these need to be trimmed with trim galore. /tools/TrimGalore/trim_galore ERR1091914.fastq
Once you have trimmed reads you can map them to a reference genome or assemble them. Lets stick with read mapping for the moment as assembly requires a bit more software downloads like megahit.
You'll want to download the reference they mention in the paper from NCBI.
Click on the genome tab in blue and you will get a .fna file. This will need to indexed with bwa.
bwa mem index SARs-CoV-2_GCF_000864885.1_ViralProj15500_genomic.fna
Next step is to map these reads with bwa mem on 4 threads and to pipe this into samtools so you get a .bam file out of the mapping.
Copy both of these files to a local drive that has IGV installed.
Bingo... There are indeed SARs-CoV-1 sequences in the listed samples in the paper. While it may be from heavily censored China, it is reproducible. ncbi.nlm.nih.gov/sra/?term=ERR1…
• • •
Missing some Tweet in this thread? You can try to
force a refresh
I like mice…
So does some lab that derived Omicron.
Very good read.
You’ll see how Molnupiravir could not have done this alone.
Random C->U mutations are not elevated in omicron but GC-AT are different in early omicron Vs post outbreak.
Mice host!
GC-AT mutations are a signature of ROS damage that makes 8-OxoG.
ROS is elevated in febrile viral infections.
These authors looked at early omicron Vs post outbreak omicron and see a difference here suggesting omicron had a different host at some point.
Mice fit the bill.
Another kind cat sent me this paper.
Another important point is the Ka/Ks analysis they do (dN/dS).
This shows 6.6X higher selective pressure on spike but not on the rest of the SARs genome.
Mice + selection against spike (vax of MAbs) could do this.
2)pseudouridine and N1-methyl should be better spelled out. Xia et al conflates the two and we will spell this out more.
PseudoU wobbles more the methylpseudoU but both significantly alter Tm and thus are ‘stickier’ bases than disrupt translation.
Very thorough.
Not only applied a live-dead like PCR looking at sgRNA, they also cultured the virus and looked for immune histochemistry confirmation of proteins.
Great chat with @jjcouey
I have some errors to confess to as I spoke to fast.
1)I was enrolled in a PhD program at UW but dropped out to focus on my job when the HGP starting racing with Craig Venter. So No PhD.
3)The vaccines have N1-Methyl Pseudouridine and I shouldn't shorthand this to Pseudouridine as the former has less wobble than the latter. N1-Methyl does alter the Tm and increase base stacking in RNA but its methyl group steals one potential H-bond.
Let’s have a look at the methods section of the paper.
2 rounds of 45 cycles of PCR?
This is what you do if you want to make polymerase slippage errors near GC rich triplet repeats.
People this library illiterate shouldn’t have the keys to $20K NovaSeq runs.
Having engineered very accurate sequencers, to address PCR polymerase error, you must utilize DNA barcodes that get attached to your PCR products in the first cycle of PCR. Any barcode that over amplifies through the PCR process is discounted.