Sequencing of RNase A treated Pfizer bivalent vaccines reveals paired-end sequencing evidence of circular plasmids and an inter-vial 72bp variation in the SV40 promoter.

Introduction

Previous work sequencing the Pfizer bivalent vaccines utilized a method that sequenced both the RNA and DNA in the sample. The RNA-Seq kit employed favored sequencing RNA and under estimated the DNA contamination levels in the vaccine. This made it difficult to assess the % of the plasmid DNA that was linear versus circular DNA. To address this problem, we RNase A treated the sample and sequenced the remaining DNA. RNase A digests RNA into small pieces that can be eliminated with purification. This enabled a deeper sampling of the DNA based molecules without the mRNA sequence obscuring and suppressing the depth of coverage for the plasmid vector.

After assembling and mapping these paired reads back to the reference genome we can look for reads that point across the part of the reference that should be circular. Since these sequencing libraries were known to be 250 base pairs in size, we can look for reads that map ~7,700 bases apart and point across the circularization point.

Let us expand on this point as it a very important aspect of next generation sequencing. Figure 2 is the DNA length of the RNase A treated libraries. We were very careful to make these very tight insert size libraries so we have an expected distance between the forward and reverse sequencing reads. The tighter your size distribution of molecules, the easier it is to detect deviations from that distribution.

Figure 2. Agilent Tape Station 4200 electropherogram of DNA molecules that were purified from the vaccine and enzymatically fragmented to afford Illumina sequencing. Illumina cant sequence 7kb molecules. They must be broken into smaller pieces, sequenced and then reassembled.

When the reads map to the reference and significantly deviate from this known DNA length, you know you have structural variations in your molecules compared to the reference genome you are mapping them too. NCBI has a tutorial on how to detect structural variation with short reads (Figure 3).

Figure 3. Depiction of short reads mapping to a reference genome. Reads can differ in anticipated insert size but also differ in the read orientation. Different read orientations help to identify inversions as seen in green or D.

This is a field the team at Agencourt Personal Genomics pioneered as they were the first Next Generation sequencing platform (SOLiD sequencing) that used TypeIIs restriction enzymes to make paired end libraries that spanned 10-20kb in size. These large insert libraries could detect novel structural variations in human genomes (Figure 4).

Figure 4. Reproduced from McKernan et al. Detection of large deletions with insert size deviations after read mapping.

The Broad Institute (the authors of IGV:Integrated Genome Viewer used in Figure 1), have a great tutorial that covers read pairing with Illumina and SOLiD sequencing.

Figure 5 Illustrates Pfizer’s disclosed vector map (EMA documentation below) on the left. The vector on the right is what the RNase A sequencing and assembly produced. The middle plasmid is a depiction of the circularized read pairs from the Top track. The top track is a depiction of a linear plasmid assembly and the reads with in that assembly that are evidence of the circular state of the molecule. The fact that the green reads are pointing off the edges of the reference and we know they are 250bp molecules, leads to one structural solution as circular molecules. The origin of replication has higher coverage because these plasmids were isolated from E.coli in mid log scale growth (EMA document) and the assembly is capturing the fact that a population of the plasmids are in mid replication.

Figure 5. Paired end reads that point outwards (outties) on a linear reference represents 250bp reads that span the noon mark on the plasmid vector. The noon mark also contains the restriction enzyme that is used for linearization (Eam1104i). It also contains the Poly A tract which varies by 14 bases from Pfizers claim. This variation is a result of PCR slippage over the Poly A.

Background: The EMA

Many thanks to

Sasha Latypova

for forwarding this EMA document. Note, on page 69, the regulators begin asking some pointy questions on the spotty genomics used to QC these products. The linearization reaction used to cut the plasmid with a restriction enzyme, is never quantitated.

Figure 6. EMA documentation critiquing Pfizers methods

Other interesting details are disclosed regarding their process. We need a mechanism to monitor dsRNA as the Immunoassay methods they are using for this have very little literature on N1-methylpseudoU (m1Ψ) . If readers find any, please post it. Counter to the poorly described immunoassay in the EMA document, we have an RNA-seq strandedness analysis that demonstrates discrete regions in the mRNA that have more anti-sense alignments than sense alignments with the sequencing reads.

Several examples of concerns raised by the EMA that could lead to a failure to detect the high levels of dsDNA or dsRNA contamination in these vials.

Figure 7. m1Ψ plays a role in the amount of dsRNA that is produced.

This is an artifact of m1Ψ being a very sticky base. Note, replacing just 4 uracils with m1Ψ jumps the Tm nearly C. Change a 1000 and you have molecular velcro.

Figure 8. Figure reproduced from Parr et al demonstrating the extreme alterations in melting temperature with m1Ψ substitution.

Pfizer’s sequencing QC work omits discussion of paired reads and if any reads bridge the cut site (Eam1104I). This cut site is adjacent to the Poly A tail and the above analysis demonstrates many reads which bridge this cut site. That can only happen on molecules that fail to cut and there are many present.

Figure 9. EMA disclosure on Pfizers sequencing methods leave more questions than answers. Where is the sequence. Why no paired end analysis?

They claim to test for background DNA but are called out for providing no data nor any methods to back those claims.

Figure 10. Trust us. We DNase’d it. Pinky sware.

This is also a classic magic trick that escapes many reviewers as they fail to appreciate that RT-qPCR amplifies both RNA and DNA. They are targeting spike with RT-qPCR. The spike sequence exists in both the plasmid vector and the spike mRNA. RT-qPCR amplifies both DNA and RNA so this assay will conflate the two.

Figure 11. The regulatory may not be aware that this RT-qPCR does NOT only amplify RNA. It also amplifies DNA.

The regulators call them out for not disclosing the CT value. Notice they also don’t like CTs over 32 but don’t disclose the CT their samples produce.

Figure 12. Regulators asking some pointy questions about their vague CT scores.

The same goes for their qPCR looking at DNA. They don’t address linearity versus circles with this.

Figure 13. Regulators noting Pfizers assays dont address linear vs circular DNA

The residual dsDNA with these undisclosed methods also varies 10X.

Figure 14. Note the 211 on the lower right. a 10X jump on just 5 batches.

More evidence that their DNase I step is in question.

Figure 15. Pfizer admitting to DNase I variability.

Results

With the mRNA removed from the sequencing, we now have a much cleaner picture of the vector sequence and must address a correction. Previous assemblies provided very short polyA tracts leading us to speculate that they must be adding the 100bp poly A on after expression from this plasmid. The RNase A libraries change this conclusion as long poly A tracts are in fact now in the vector. This is a PANDA presentation on the data before we had this information. Note, my hypothesis regarding their ligation of poly A tails on after T7 polymerization has been falsified with the RNase A libraries.

PANDA @PanData19

More public info on the liability-free Covid vaccine ingredients is imperative. The EMA & TGA note fragmented RNA & smeary western blots, suggesting the vaccine production lacks fidelity and transparency. @Kevin_McKernan presents his technical findings: bit.ly/3L0a2iS /1

The Pfizer vaccine sequence is now 7810 bases. The Pfizer disclosure is just 14 bases longer (7824bp). The disagreement is in the linker in the Poly A tail. This is a 30A→10bp linker→ 70A region described by Nance et al. It is very likely Illumina sequencing exhibits polymerase slippage over this region and the disagreement is artifactual.

Figure 16. Samtools stats for the BAM file on the left. RNase treated Pfizer bivalent vial number 1 contig k141_107. This sequence is 14bases short of Pfizers depiction.

>k141_107 flag=1 multi=7745.0000 len=7810

TTTTTTTTTTTTTTTTTTTTTTTTTTTTTTTTGCTAGCTCCAGGGTGTGGCTGGCACGAAATTGACCAACCCTGGGGTTAGTATAGCTTAGTTAAACTTTCGTTTATTGCTAAAGGTTAATCACTGCTGTTTCCCGTGGGGGTGTGGCTAGGCTAAGCGTTTTGAGCTGCATTGCTGCGTGCTTGGGAGGTGTCTGGAACTAGCAGAGGTGGTGAGTGGGGCAGGTGGAGGTGGGAGCATACCTGGGACCCGAGGTCGGGGGAGACTCGGGGTACCCAGGACGGGAAAGGGGCAGCTAGCATTGCGTGCATGCAGTACCAGCTCGAGTCATCATGTGTAGTGCAGTTTCACGCCCTTCAGCACGGGCTCAGAATCGTCCTCGTCGAACTTGCAGCAGCTGCCACAGCTACAACAGCCCTTCAGGCAGCTACAGCAGCTGGTCATGCAACACAGCATGATTGTGACCATCACGATGGCAATCAGTCCGGCGATAAAGCCCAGCCAGATGTACCAGGGCCACTTGATGTACTGCTCGTACTTCCCCAGTTCTTGCAGGTCGATCAGGCTCTCGTTCAGATTCTTGGCCACCTCGTTCAGCCGGTCGATCTCTTTCTGGATGTTCACGACGCTGGCATTGATTCCGCTGATATCGCCCAGGTCCACGTCGGGGCTTGTGTGGTTCTTAAAGTACTTGTCCAGTTCCTCTTTGAAGCTGTCCAGCTCGGGCTGCAGAGGGTCGTACACGGTATTGTTCACAATGCCGATCACGACGTCGCAGTTGCCAGACACGAAGGTGTTGTCGGTGGTGATGATCTGGGGCTCGTAGAAGTTCCGCTGTGTCACGAACCAATGGGTGCCGTTGGACACGAACACGCCTTCTCTAGGAAAGTGGGCTTTGCCGTCGTGGCAGATGGCTGGAGCGGTGGTGAAATTCTTCTCTTGAGCGGGCACATATGTCACGTGCAGAAACACCACGCCGTGAGGGGCAGACTGAGGGAAGCTCATCAGGTGGTAGCCCTTGCCGCAAAAGTCCACTCTCTTGCTCTGGCCCAGCACACACTCAGACATCTTGGTGGCGGCCAGATTGGCAGAGGCTCTAATCTCGGCGGCTCTGATCAGCTGCTGGGTCACGTATGTCTGGAGGCTCTGCAGTCTGCCTGTGATCAGTCTGTCGATCTGCACCTCGGCCTCAGGAGGGTCCAGTCTGCTCAGGATATCGTTCAGCACAGAGCTGATGGCGCCGAACTTGGAGGACAGCTGCTTGACCAGGGTGTTCAGTGCCTGGGCATTGTGGTTGACCACGTCCTGCAGCTTTCCCAGGGCGCTTGCTGTGCTGCTCAGGCTGTCCTGGATCTTGCCGATGGCGCTGTTGAACTGGTTGGCGATCAGCTTCTGGTTCTCGTACAGCACATTCTGGGTCACTCCGATGCCGTTGAACCGGTAGGCCATCTGCATAGCAAAGGGGATCTGCAGAGCGGCGCCTGCTCCAAATGTCCAGCCGCTTGTGATTGTGCCGGCCAGCAGGGCAGATGTGTACTGGGCGATCATCTCATCGGTCAGCAGAGGAGGCAGCACTGTCAGTCCGTTAAACTTCTGGGCGCAAATCAGATCCCTGGCGGCAATGTCGCCCAGACAATCGCCATACTGCTTGATGAAGCCGGCGTCGGCCAGTGTCACTTTGTTGAACAGCAGGTCCTCGATGAAGCTCCGCTTGCTGGGCTTGCTAGGATCGGGCAGAATCTGGCTGAAATTGAAGCCGCCGAAGTACTTGATAGGAGGGGTCTTGTAGATCTGCTTCACTTGGGCGAACACCTCTTGGGTGTTCTTGTCCTGTTCCACGGCGATCCCTGTCAGGGCTCTTTTCAGCTGGGTGCAGAAGCTGCCGTACTGCAGCAGCAGGTTGGAGCACTCGGTGGAATCGCCGCAGATGTACATGGTGCAGTCCACGCTGGTCTTGGTCATGGACACAGGCAGGATCTCTGTGGTCACGCTGATGGTGAAGTTGGTGGGGATAGCGATAGAGTTGTTGGAGTAGGCCACGCTGTTCTCGGCGCCCAGAGACATTGTGTAGGCAATGATGCTCTGGCTGGCCACGCTTCTGGCTCTCCGGTGGCTCTTTGTCTGTGTCTGGTAGCTGGCGCAGATTCCAGCGCCGATGGGGATGTCGCACTCGTAGCTATTGTTCACGTACTCGGCTCCGATCAGACAGCCGGCTCTGGTCTGAAACACATTGCTGCCGGTGGAGTACACCCGCCATGTAGGTGTCAGCTGATCGGCGTGAATGGCCACGGGCACTTCGGTACAGTTCACGCCCTGGTACAGCACTGCCACCTGATTGCTGGTGTTGGTGCCAGGGGTGATCACAGACACTCCGCCGAAGCTGCAAGGGGTGATGTCCAGGATTTCCAGTGTCTGGGGATCTCTAACGGCGTCTGTGGTATCGGCGATATCCCGGCCAAACTGCTGGAATGGCAGGAACTTCTTGTTGCTCTCTGTCAGCACGCCGGTGCCGGTCAGGCCGTTGAAGTTGAAGTTCACGCATTTGTTCTTCACGAGATTGGTGCTTTTCTTAGGGCCGCACACTGTGGCAGGGGCATGCAGCAGTTCGAAGCTCAGCACCACCACTCTGTAGGGCTGGTGGCCCACGCCGTATGTGGGCCTAAAGCCGTAGGACTGCAGTGGGAAGTAGCAGTTCACGCCTGCCACGCCGTTACAAGGCTTGTTGCCGGCCTGATAGATCTCGGTGGAGATGTCCCGCTCGAAGGGCTTCAGATTGGACTTCCGGAACAGCCGGTACCTGTAATTGTAGTTGCCGCCGACTTTGGAGTCCAGCTTGTTGCTGTTCCAGGCAATCACACAGCCGGTGAAGTCGTCGGGCAGCTTGTAGTTGTAGTCGGCGATGTTGCCTGTCTGTCCAGGGGCAATCTGCCGCACTTCGTTTCCCCGGATCACGAAGCTGTCGGCGTACACGTTTGTGAAGCACAGGTCGTTCAGCTTGGTAGGGGACACGCCGTAGCACTTGAATGCGAAGAAGGGGGCGAAGTTGTACAGCACGGAGTAGTCGGCCACGCAATTGCTGATCCGCTTCCGGTTCCAGGCGTACACAGAGGCGAATCTGGTGGCATTGAACACCTCGTCGAAGGGGCACAGATTGGTGATATTGGGGAACCGCACGATGGATTCGGTGGGCTGCACCCGGAAGTTGCTGGTCTGGTAGATGCCCTTTTCCACGGTGAAGGACTTCAGGGTGCACTTTGTCTCGCTCAGAGGATCCAGAGCACAATCCACGGCGTCGGTGATGGTGCCGTTCTCGTTGTACTTCAGCAGGAAGGTTCTAGGCTGCAGGTAGCCCACATAGTAAGCGGCGGCACCAGCTGTCCATCCGCTGCTGCTATCGCCAGGTGTCAGGTAGCTTCTGTGCAGGGCCAGCAGTGTCTGAAACCGGGTGATGTTGATGCCGATGGGCAGATCCACCAGGGGTTCCAGAGCAGAGAAGCCCTGAGGCAGATCCCGGCCGAGGTTGATAGGGGTGTGCTTGCTGTAGATCTTGAAGTAGCCGTCGATGTTCTTAAACACGAACTCGCGCAGGTTCTTGAAGTTGCCCTGCTTGCCTTCCAGGTCCATCAGGAAAGGCTGGGACACGTACTCGAAGGTGCAGTTGTTGGCGCTGCTGTACACCCGGAACTCGCTTTCCATCCAGCTCTTGTTGTTCTTGTGGTAGTAGACGTCCAGGAAGGGGTCGTTGCAGAACTGGAACTCGCACACTTTGATGACCACGTTGGTGGCGTTGTTCACGATCAGCAGGCTCTGGGTCTTGCTGTCCAGTGTGGTGCCGAAGATCCAGCCTCTGATGATGTTGGACTTCTCGGTGCTGGCAAAGTACACCCCGTCGTTGAAGGGCAGCACGGGGTTGTCGAATCTCTTGGTGCCATTGGTGCCGGAGATGGCGTGGAACCAGGTCACGTTGCTGAAGAAAGGCAGGAACAGGTCCTGGGTAGAGTGCAGCACGCTGGATCTGAACACCTTGTCGGGGTAGTACACGCCTCTGGTAAAGCTGTTGGTGTATGACTGTGTTCTGGTGATCAGGTTCACACACTGGCTGGACACCAGAGGCAGCAGCACCAGGAACACGAACATGGTGGCGGGTTCTCTCTGAGTCTGTGGGGACCAGAAGAATACTAGTTTATTCTTATAGTGAGTCGTATTAATTAATAACTAATGCATGGCGGTAATACGGTTATCCACAGAATCAGGGGATAACGCAGGAAAGAACATGTGAGCAAAAGGCCAGCAAAAGGCCAGGAACCGTAAAAAGGCCGCGTTGCTGGCGTTTTTCCATAGGCTCCGCCCCCCTGACGAGCATCACAAAAATCGACGCTCAAGTCAGAGGTGGCGAAACCCGACAGGACTATAAAGATACCAGGCGTTTCCCCCTGGAAGCTCCCTCGTGCGCTCTCCTGTTCCGACCCTGCCGCTTACCGGATACCTGTCCGCCTTTCTCCCTTCGGGAAGCGTGGCGCTTTCTCATAGCTCACGCTGTAGGTATCTCAGTTCGGTGTAGGTCGTTCGCTCCAAGCTGGGCTGTGTGCACGAACCCCCCGTTCAGCCCGACCGCTGCGCCTTATCCGGTAACTATCGTCTTGAGTCCAACCCGGTAAGACACGACTTATCGCCACTGGCAGCAGCCACTGGTAACAGGATTAGCAGAGCGAGGTATGTAGGCGGTGCTACAGAGTTCTTGAAGTGGTGGCCTAACTACGGCTACACTAGAAGAACAGTATTTGGTATCTGCGCTCTGCTGAAGCCAGTTACCTTCGGAAAAAGAGTTGGTAGCTCTTGATCCGGCAAACAAACCACCGCTGGTAGCGGTGGTTTTTTTGTTTGCAAGCAGCAGATTACGCGCAGAAAAAAAGGATCTCAAGAAGATCCTTTGATCTTTTCTACGGGGTCTGACGCTCAGTGGAACGAAAACTCACGTTAAGGGATTTTGGTCATGAGATTATCAAAAAGGATCTTCACCTAGATCCTTTTAAATTAAAAATGAAGTTTTAAATCAATCTAAAGTATATATGAGTAACCTGAGGCTATGGCAGGGCCTGCCGCCCCGACGTTGGCTGCGAGCCCTGGGCCTTCACCCGAACTTGGGGGGTGGGGTGGGGAAAAGGAAGAAACGCGGGCGTATTGGCCCCAATGGGGTCTCGGTGGGGTATCGACAGAGTGCCAGCCCTGGGACCGAACCCCGCGTTTATGAACAAACGACCCAACACCGTGCGTTTTATTCTGTCTTTTTATTGCCGTCATAGCGCGGGTTCCTTCCGGTATTGTCTCCTTCCGTGTTTCAGTTAGCCTCCCCCTAGGGTGGGCGAAGAACTCCAGCATGAGATCCCCGCGCTGGAGGATCATCCAGCCGGCGTCCCGGAAAACGATTCCGAAGCCCAACCTTTCATAGAAGGCGGCGGTGGAATCGAAATCTCGTGATGGCAGGTTGGGCGTCGCTTGGTCGGTCATTTCGAACCCCAGAGTCCCGCTCAGAAGAACTCGTCAAGAAGGCGATAGAAGGCGATGCGCTGCGAATCGGGAGCGGCGATACCGTAAAGCACGAGGAAGCGGTCAGCCCATTCGCCGCCAAGTTCTTCAGCAATATCACGGGTAGCCAACGCTATGTCCTGATAGCGGTCCGCCACACCCAGCCGGCCACAGTCGATGAATCCAGAAAAGCGGCCATTTTCCACCATGATATTCGGCAAGCAGGCATCGCCATGGGTCACGACGAGATCCTCGCCGTCGGGCATGCTCGCCTTGAGCCTGGCGAACAGTTCGGCTGGCGCGAGCCCCTGATGTTCTTCGTCCAGATCATCCTGATCGACAAGACCGGCTTCCATCCGAGTACGTGCTCGCTCGATGCGATGTTTCGCTTGGTGGTCGAATGGGCAGGTAGCCGGATCAAGCGTATGCAGCCGCCGCATTGCATCAGCCATGATGGATACTTTCTCGGCAGGAGCAAGGTGAGATGACAGGAGATCCTGCCCCGGCACTTCGCCCAATAGCAGCCAGTCCCTTCCCGCTTCAGTGACAACGTCGAGCACAGCTGCGCAAGGAACGCCCGTCGTGGCCAGCCACGATAGCCGCGCTGCCTCGTCTTGCAGTTCATTCAGGGCACCGGACAGGTCGGTCTTGACAAAAAGAACCGGGCGCCCCTGCGCTGACAGCCGGAACACGGCGGCATCAGAGCAGCCGATTGTCTGTTGTGCCCAGTCATAGCCGAATAGCCTCTCCACCCAAGCGGCCGGAGAACCTGCGTGCAATCCATCTTGTTCAATCATGCGAAACGATCCTCATCCTGTCTCTTGATCGATCTTTGCAAAAGCCTAGGCCTCCAAAAAAGCCTCCTCACTACTTCTGGAATAGCTCAGAGGCCGAGGCGGCCTCGGCCTCTGCATAAATAAAAAAAATTAGTCAGCCATGGGGCGGAGAATGGGCGGAACTGGGCGGAGTTAGGGGCGGGATGGGCGGAGTTAGGGGCGGGACTATGGTTGCTGACTAATTGAGATGCATGCTTTGCATACTTCTGCCTGCTGGGGAGCCTGGGGACTTTCCACACCTGGTTGCTGACTAATTGAGATGCATGCTTTGCATACTTCTGCCTGCTGGGGAGCCTGGGGACTTTCCACACCCTAACTGACACACATTCCACAGCTGGTTCTTTCCGCCTCAGGATTCTTCCTTTTTCAATATTATTGAAGCATTTATCAGGGTTATTGTCTCATGAGCGGATACATATTTGAATGTATTTAGAAAAATAAACAAATAGGGGTTCCGCGCACATTTCCCCGAAAAGTGCCACCTGACGCGCCCTGTAGCGGCGCATTAAGCGCGGCGGGTGTGGTGGTTACGCGCAGCGTGACCGCTACACTTGCCAGCGCCCTAGCGCCCGCTCCTTTCGCTTTCTTCCCTTCCTTTCTCGCCACGTTCGCCGGCTTTCCCCGTCAAGCTCTAAATCGGGGGCTCCCTTTAGGGTTCCGATTTAGTGCTTTACGGCACCTCGACCCCAAAAAACTTGATTAGGGTGATGGTTCACGTAGTGGGCCATCGCCCTGATAGACGGTTTTTCGCCCTTTGACGTTGGAGTCCACGTTCTTTAATAGTGGACTCTTGTTCCAAACTGGAACAACACTCAACCCTATCTCGGTCTATTCTTTTGATTTATAAGGGATTTTGCCGATTTCGGCCTATTGGTTAAAAAATGAGCTGATTTAACAAAAATTTAACGCGAATTTTAACAAAATATTAACGCTTACAATTTACGCGTTAAGATACATTGATGAGTTTGGACAAACCACAACTAGAATGCAGTGAAAAAAATGCTTTATTTGTGAAATTTGTGATGCTATTGCTTTATTTGTAACCATTATAAGCTGCAATAAACAAGTTAACAACAACAATTGCATTCATTTTATGTTTCAGGTTCAGGGGGAGGTGTGGGAGGTTTTTTAAAGCAAGTAAAACCTCTACAAATGTGGTATGGCTGATTATGATCATGAACAGACTGTGAGGACTGAGGGGCCTGAAATGAGCCTTGGGACTGTGAATCTAAAATACACAAACAATTAGAATCAGTAGTTTAACACATTATACACTTAAAAATTGGATCTCCATTCGCCATTCAGGCTGCGCAACTGTTGGGAAGGGCGATGTTAAACGGCGGAGCTACCACACCGGTTGGAGCTCTTCTTTTTTTTTTTTTTTTTTTTGTTTTTTTTTTTTTTTTTTTTTTTTTTTTTTTTTTTTTTTTTTTTTTTTTTTTTTTTTT

Figure 17. Samtools stats on Pbiv2_RNase_WM_K141_23 bam file on the left. Sequencing of the RNase A treated Pfizer bivalent vial 2 delivered a vector with a 72bp deletion in the SV40 promoter.

>k141_23 flag=1 multi=6914.0000 len=7667

TTTTTTTTTTTTTTTTTTTTGCTAGCTCCAGGGTGTGGCTGGCACGAAATTGACCAACCCTGGGGTTAGTATAGCTTAGTTAAACTTTCGTTTATTGCTAAAGGTTAATCACTGCTGTTTCCCGTGGGGGTGTGGCTAGGCTAAGCGTTTTGAGCTGCATTGCTGCGTGCTTGGGAGGTGTCTGGAACTAGCAGAGGTGGTGAGTGGGGCAGGTGGAGGTGGGAGCATACCTGGGACCCGAGGTCGGGGGAGACTCGGGGTACCCAGGACGGGAAAGGGGCAGCTAGCATTGCGTGCATGCAGTACCAGCTCGAGTCATCATGTGTAGTGCAGTTTCACGCCCTTCAGCACGGGCTCAGAATCGTCCTCGTCGAACTTGCAGCAGCTGCCACAGCTACAACAGCCCTTCAGGCAGCTACAGCAGCTGGTCATGCAACACAGCATGATTGTGACCATCACGATGGCAATCAGTCCGGCGATAAAGCCCAGCCAGATGTACCAGGGCCACTTGATGTACTGCTCGTACTTCCCCAGTTCTTGCAGGTCGATCAGGCTCTCGTTCAGATTCTTGGCCACCTCGTTCAGCCGGTCGATCTCTTTCTGGATGTTCACGACGCTGGCATTGATTCCGCTGATATCGCCCAGGTCCACGTCGGGGCTTGTGTGGTTCTTAAAGTACTTGTCCAGTTCCTCTTTGAAGCTGTCCAGCTCGGGCTGCAGAGGGTCGTACACGGTATTGTTCACAATGCCGATCACGACGTCGCAGTTGCCAGACACGAAGGTGTTGTCGGTGGTGATGATCTGGGGCTCGTAGAAGTTCCGCTGTGTCACGAACCAATGGGTGCCGTTGGACACGAACACGCCTTCTCTAGGAAAGTGGGCTTTGCCGTCGTGGCAGATGGCTGGAGCGGTGGTGAAATTCTTCTCTTGAGCGGGCACATATGTCACGTGCAGAAACACCACGCCGTGAGGGGCAGACTGAGGGAAGCTCATCAGGTGGTAGCCCTTGCCGCAAAAGTCCACTCTCTTGCTCTGGCCCAGCACACACTCAGACATCTTGGTGGCGGCCAGATTGGCAGAGGCTCTAATCTCGGCGGCTCTGATCAGCTGCTGGGTCACGTATGTCTGGAGGCTCTGCAGTCTGCCTGTGATCAGTCTGTCGATCTGCACCTCGGCCTCAGGAGGGTCCAGTCTGCTCAGGATATCGTTCAGCACAGAGCTGATGGCGCCGAACTTGGAGGACAGCTGCTTGACCAGGGTGTTCAGTGCCTGGGCATTGTGGTTGACCACGTCCTGCAGCTTTCCCAGGGCGCTTGCTGTGCTGCTCAGGCTGTCCTGGATCTTGCCGATGGCGCTGTTGAACTGGTTGGCGATCAGCTTCTGGTTCTCGTACAGCACATTCTGGGTCACTCCGATGCCGTTGAACCGGTAGGCCATCTGCATAGCAAAGGGGATCTGCAGAGCGGCGCCTGCTCCAAATGTCCAGCCGCTTGTGATTGTGCCGGCCAGCAGGGCAGATGTGTACTGGGCGATCATCTCATCGGTCAGCAGAGGAGGCAGCACTGTCAGTCCGTTAAACTTCTGGGCGCAAATCAGATCCCTGGCGGCAATGTCGCCCAGACAATCGCCATACTGCTTGATGAAGCCGGCGTCGGCCAGTGTCACTTTGTTGAACAGCAGGTCCTCGATGAAGCTCCGCTTGCTGGGCTTGCTAGGATCGGGCAGAATCTGGCTGAAATTGAAGCCGCCGAAGTACTTGATAGGAGGGGTCTTGTAGATCTGCTTCACTTGGGCGAACACCTCTTGGGTGTTCTTGTCCTGTTCCACGGCGATCCCTGTCAGGGCTCTTTTCAGCTGGGTGCAGAAGCTGCCGTACTGCAGCAGCAGGTTGGAGCACTCGGTGGAATCGCCGCAGATGTACATGGTGCAGTCCACGCTGGTCTTGGTCATGGACACAGGCAGGATCTCTGTGGTCACGCTGATGGTGAAGTTGGTGGGGATAGCGATAGAGTTGTTGGAGTAGGCCACGCTGTTCTCGGCGCCCAGAGACATTGTGTAGGCAATGATGCTCTGGCTGGCCACGCTTCTGGCTCTCCGGTGGCTCTTTGTCTGTGTCTGGTAGCTGGCGCAGATTCCAGCGCCGATGGGGATGTCGCACTCGTAGCTATTGTTCACGTACTCGGCTCCGATCAGACAGCCGGCTCTGGTCTGAAACACATTGCTGCCGGTGGAGTACACCCGCCATGTAGGTGTCAGCTGATCGGCGTGAATGGCCACGGGCACTTCGGTACAGTTCACGCCCTGGTACAGCACTGCCACCTGATTGCTGGTGTTGGTGCCAGGGGTGATCACAGACACTCCGCCGAAGCTGCAAGGGGTGATGTCCAGGATTTCCAGTGTCTGGGGATCTCTAACGGCGTCTGTGGTATCGGCGATATCCCGGCCAAACTGCTGGAATGGCAGGAACTTCTTGTTGCTCTCTGTCAGCACGCCGGTGCCGGTCAGGCCGTTGAAGTTGAAGTTCACGCATTTGTTCTTCACGAGATTGGTGCTTTTCTTAGGGCCGCACACTGTGGCAGGGGCATGCAGCAGTTCGAAGCTCAGCACCACCACTCTGTAGGGCTGGTGGCCCACGCCGTATGTGGGCCTAAAGCCGTAGGACTGCAGTGGGAAGTAGCAGTTCACGCCTGCCACGCCGTTACAAGGCTTGTTGCCGGCCTGATAGATCTCGGTGGAGATGTCCCGCTCGAAGGGCTTCAGATTGGACTTCCGGAACAGCCGGTACCTGTAATTGTAGTTGCCGCCGACTTTGGAGTCCAGCTTGTTGCTGTTCCAGGCAATCACACAGCCGGTGAAGTCGTCGGGCAGCTTGTAGTTGTAGTCGGCGATGTTGCCTGTCTGTCCAGGGGCAATCTGCCGCACTTCGTTTCCCCGGATCACGAAGCTGTCGGCGTACACGTTTGTGAAGCACAGGTCGTTCAGCTTGGTAGGGGACACGCCGTAGCACTTGAATGCGAAGAAGGGGGCGAAGTTGTACAGCACGGAGTAGTCGGCCACGCAATTGCTGATCCGCTTCCGGTTCCAGGCGTACACAGAGGCGAATCTGGTGGCATTGAACACCTCGTCGAAGGGGCACAGATTGGTGATATTGGGGAACCGCACGATGGATTCGGTGGGCTGCACCCGGAAGTTGCTGGTCTGGTAGATGCCCTTTTCCACGGTGAAGGACTTCAGGGTGCACTTTGTCTCGCTCAGAGGATCCAGAGCACAATCCACGGCGTCGGTGATGGTGCCGTTCTCGTTGTACTTCAGCAGGAAGGTTCTAGGCTGCAGGTAGCCCACATAGTAAGCGGCGGCACCAGCTGTCCATCCGCTGCTGCTATCGCCAGGTGTCAGGTAGCTTCTGTGCAGGGCCAGCAGTGTCTGAAACCGGGTGATGTTGATGCCGATGGGCAGATCCACCAGGGGTTCCAGAGCAGAGAAGCCCTGAGGCAGATCCCGGCCGAGGTTGATAGGGGTGTGCTTGCTGTAGATCTTGAAGTAGCCGTCGATGTTCTTAAACACGAACTCGCGCAGGTTCTTGAAGTTGCCCTGCTTGCCTTCCAGGTCCATCAGGAAAGGCTGGGACACGTACTCGAAGGTGCAGTTGTTGGCGCTGCTGTACACCCGGAACTCGCTTTCCATCCAGCTCTTGTTGTTCTTGTGGTAGTAGACGTCCAGGAAGGGGTCGTTGCAGAACTGGAACTCGCACACTTTGATGACCACGTTGGTGGCGTTGTTCACGATCAGCAGGCTCTGGGTCTTGCTGTCCAGTGTGGTGCCGAAGATCCAGCCTCTGATGATGTTGGACTTCTCGGTGCTGGCAAAGTACACCCCGTCGTTGAAGGGCAGCACGGGGTTGTCGAATCTCTTGGTGCCATTGGTGCCGGAGATGGCGTGGAACCAGGTCACGTTGCTGAAGAAAGGCAGGAACAGGTCCTGGGTAGAGTGCAGCACGCTGGATCTGAACACCTTGTCGGGGTAGTACACGCCTCTGGTAAAGCTGTTGGTGTATGACTGTGTTCTGGTGATCAGGTTCACACACTGGCTGGACACCAGAGGCAGCAGCACCAGGAACACGAACATGGTGGCGGGTTCTCTCTGAGTCTGTGGGGACCAGAAGAATACTAGTTTATTCTTATAGTGAGTCGTATTAATTAATAACTAATGCATGGCGGTAATACGGTTATCCACAGAATCAGGGGATAACGCAGGAAAGAACATGTGAGCAAAAGGCCAGCAAAAGGCCAGGAACCGTAAAAAGGCCGCGTTGCTGGCGTTTTTCCATAGGCTCCGCCCCCCTGACGAGCATCACAAAAATCGACGCTCAAGTCAGAGGTGGCGAAACCCGACAGGACTATAAAGATACCAGGCGTTTCCCCCTGGAAGCTCCCTCGTGCGCTCTCCTGTTCCGACCCTGCCGCTTACCGGATACCTGTCCGCCTTTCTCCCTTCGGGAAGCGTGGCGCTTTCTCATAGCTCACGCTGTAGGTATCTCAGTTCGGTGTAGGTCGTTCGCTCCAAGCTGGGCTGTGTGCACGAACCCCCCGTTCAGCCCGACCGCTGCGCCTTATCCGGTAACTATCGTCTTGAGTCCAACCCGGTAAGACACGACTTATCGCCACTGGCAGCAGCCACTGGTAACAGGATTAGCAGAGCGAGGTATGTAGGCGGTGCTACAGAGTTCTTGAAGTGGTGGCCTAACTACGGCTACACTAGAAGAACAGTATTTGGTATCTGCGCTCTGCTGAAGCCAGTTACCTTCGGAAAAAGAGTTGGTAGCTCTTGATCCGGCAAACAAACCACCGCTGGTAGCGGTGGTTTTTTTGTTTGCAAGCAGCAGATTACGCGCAGAAAAAAAGGATCTCAAGAAGATCCTTTGATCTTTTCTACGGGGTCTGACGCTCAGTGGAACGAAAACTCACGTTAAGGGATTTTGGTCATGAGATTATCAAAAAGGATCTTCACCTAGATCCTTTTAAATTAAAAATGAAGTTTTAAATCAATCTAAAGTATATATGAGTAACCTGAGGCTATGGCAGGGCCTGCCGCCCCGACGTTGGCTGCGAGCCCTGGGCCTTCACCCGAACTTGGGGGGTGGGGTGGGGAAAAGGAAGAAACGCGGGCGTATTGGCCCCAATGGGGTCTCGGTGGGGTATCGACAGAGTGCCAGCCCTGGGACCGAACCCCGCGTTTATGAACAAACGACCCAACACCGTGCGTTTTATTCTGTCTTTTTATTGCCGTCATAGCGCGGGTTCCTTCCGGTATTGTCTCCTTCCGTGTTTCAGTTAGCCTCCCCCTAGGGTGGGCGAAGAACTCCAGCATGAGATCCCCGCGCTGGAGGATCATCCAGCCGGCGTCCCGGAAAACGATTCCGAAGCCCAACCTTTCATAGAAGGCGGCGGTGGAATCGAAATCTCGTGATGGCAGGTTGGGCGTCGCTTGGTCGGTCATTTCGAACCCCAGAGTCCCGCTCAGAAGAACTCGTCAAGAAGGCGATAGAAGGCGATGCGCTGCGAATCGGGAGCGGCGATACCGTAAAGCACGAGGAAGCGGTCAGCCCATTCGCCGCCAAGTTCTTCAGCAATATCACGGGTAGCCAACGCTATGTCCTGATAGCGGTCCGCCACACCCAGCCGGCCACAGTCGATGAATCCAGAAAAGCGGCCATTTTCCACCATGATATTCGGCAAGCAGGCATCGCCATGGGTCACGACGAGATCCTCGCCGTCGGGCATGCTCGCCTTGAGCCTGGCGAACAGTTCGGCTGGCGCGAGCCCCTGATGTTCTTCGTCCAGATCATCCTGATCGACAAGACCGGCTTCCATCCGAGTACGTGCTCGCTCGATGCGATGTTTCGCTTGGTGGTCGAATGGGCAGGTAGCCGGATCAAGCGTATGCAGCCGCCGCATTGCATCAGCCATGATGGATACTTTCTCGGCAGGAGCAAGGTGAGATGACAGGAGATCCTGCCCCGGCACTTCGCCCAATAGCAGCCAGTCCCTTCCCGCTTCAGTGACAACGTCGAGCACAGCTGCGCAAGGAACGCCCGTCGTGGCCAGCCACGATAGCCGCGCTGCCTCGTCTTGCAGTTCATTCAGGGCACCGGACAGGTCGGTCTTGACAAAAAGAACCGGGCGCCCCTGCGCTGACAGCCGGAACACGGCGGCATCAGAGCAGCCGATTGTCTGTTGTGCCCAGTCATAGCCGAATAGCCTCTCCACCCAAGCGGCCGGAGAACCTGCGTGCAATCCATCTTGTTCAATCATGCGAAACGATCCTCATCCTGTCTCTTGATCGATCTTTGCAAAAGCCTAGGCCTCCAAAAAAGCCTCCTCACTACTTCTGGAATAGCTCAGAGGCCGAGGCGGCCTCGGCCTCTGCATAAATAAAAAAAATTAGTCAGCCATGGGGCGGAGAATGGGCGGAACTGGGCGGAGTTAGGGGCGGGATGGGCGGAGTTAGGGGCGGGACTATGGTTGCTGACTAATTGAGATGCATGCTTTGCATACTTCTGCCTGCTGGGGAGCCTGGGGACTTTCCACACCCTAACTGACACACATTCCACAGCTGGTTCTTTCCGCCTCAGGATTCTTCCTTTTTCAATATTATTGAAGCATTTATCAGGGTTATTGTCTCATGAGCGGATACATATTTGAATGTATTTAGAAAAATAAACAAATAGGGGTTCCGCGCACATTTCCCCGAAAAGTGCCACCTGACGCGCCCTGTAGCGGCGCATTAAGCGCGGCGGGTGTGGTGGTTACGCGCAGCGTGACCGCTACACTTGCCAGCGCCCTAGCGCCCGCTCCTTTCGCTTTCTTCCCTTCCTTTCTCGCCACGTTCGCCGGCTTTCCCCGTCAAGCTCTAAATCGGGGGCTCCCTTTAGGGTTCCGATTTAGTGCTTTACGGCACCTCGACCCCAAAAAACTTGATTAGGGTGATGGTTCACGTAGTGGGCCATCGCCCTGATAGACGGTTTTTCGCCCTTTGACGTTGGAGTCCACGTTCTTTAATAGTGGACTCTTGTTCCAAACTGGAACAACACTCAACCCTATCTCGGTCTATTCTTTTGATTTATAAGGGATTTTGCCGATTTCGGCCTATTGGTTAAAAAATGAGCTGATTTAACAAAAATTTAACGCGAATTTTAACAAAATATTAACGCTTACAATTTACGCGTTAAGATACATTGATGAGTTTGGACAAACCACAACTAGAATGCAGTGAAAAAAATGCTTTATTTGTGAAATTTGTGATGCTATTGCTTTATTTGTAACCATTATAAGCTGCAATAAACAAGTTAACAACAACAATTGCATTCATTTTATGTTTCAGGTTCAGGGGGAGGTGTGGGAGGTTTTTTAAAGCAAGTAAAACCTCTACAAATGTGGTATGGCTGATTATGATCATGAACAGACTGTGAGGACTGAGGGGCCTGAAATGAGCCTTGGGACTGTGAATCTAAAATACACAAACAATTAGAATCAGTAGTTTAACACATTATACACTTAAAAATTGGATCTCCATTCGCCATTCAGGCTGCGCAACTGTTGGGAAGGGCGATGTTAAACGGCGGAGCTACCACACCGGTTGGAGCTCTTCTTTTTTTTTTTTTTTTTTTT

These two consensus sequences differ slightly in their length of the poly A tail but most strikingly with a tandem duplication in the SV40 promoter of Vial 1 compared to Vial 2.

Figure 18. Alignment of the vectors from the different vials of the same lot reveals a 72 base insertion in Vial 1.

This is a well known way to turbo SV40. Sampling just two Pfizer vials delivers two different expression vectors? These vials are from the same lot. The inserts are 100% identical and unrelated to bivalent vaccines. I wonder how many more plasmids are out there?

The 72bp repeat is known to facilitate nuclear import.

Conclusions

1)Paired end reads support the existence of circular plasmids in the Pfizer bivalent vaccines. This is supported by transformation in kanamycin sensitive E.coli. The exact ratio of circular versus linear DNA is still being evaluated but the levels of DNA contamination exceed the EMA specifications by several orders of magnitude.

2)EMA documentation reveals regulators are unsatisfied with the molecular methods used to evaluate linearization. There are concerns over non specific CT values (<32) in the PCR methods used for quantification. Additional concerns are voiced over the variation in DNase I activity and methods used to evaluate its performance.

3)A long poly A tract is evident in the vectors reducing the odds of off target poly adenylation. The 10bp linker isn’t well resolved with Illumina sequencing and the two vectors also differ in this artifactual regard.

4)A known SV40 promoter improvement is evident in 1/2 of the vials from the same lot of vaccine.

Limitations.

While RNase A is effective at removing RNA, there are conditions in which it can also eliminate DNA. This may be pertinent to how Pfizer is assessing their dsDNA contamination as they are using a single RNase and the literature on RNase A and m1Ψ hybrids with DNA is sparse. More controls with alternative RNases like RNase R will help refine the estimates of circular to linear DNA.

---

Methods

Vaccine purification, RNase A treatment and sequencing methods were published previously but is repeated here for clarity.

RNase A treatment of the Vaccines

RNase A cleaves both uracils and cytosines. N1-methylpseudouridine is known to be RNAse-L resistant but RNase A will cleave cytosines which still exist in the mRNAs. This leaves predominantly DNA for sequencing. Vaccine mRNA that was previously sequenced and discussed here, was treated at 37C for 20 minutes with 3ul of 20 Units/ul Monarch RNase A from NEB. The RNase reaction was purified using 1.5X of SenSATIVAx (Medicinal Genomics #420001). Sample were eluted in 20ul ddH20 after DNA purification. 15ul was used for DNA sequencing and 5ul used for transformation of E.coli.

Whole genome shotgun of Pre-transformation RNase’d Vaccines.

15ul of the DNA (prior to E.coli transformation) was converted into sequence ready libraries using Watchmakers Genomics WGS library construction kit.

Analysis pipeline

Reads were demultiplexed and processed with

• Trimgalore - Removes Illumina Sequencing adaptors.

• Megahit- assembles reads into contigs.

• Megahit for SARs-CoV-2

• Samtools- generates BAM files for viewing in IGV.

• Samtools stats used to calculate outie reads.

• BWA-mem- Short read mapper used to align reads back to the assembled references.

• SnapGene software- (www.snapgene.com)- Used to visualize and annotate expression vectors

• IGV- Integrated Genome Viewer used to visualize Illumina sequencing reads.

---

BAM files

contig specific BAM files were created using samtools

samtools view -h input.bam contig_name -O BAM > contig.bam; samtools index contig.bam;

Samtools stats run on a each contig in each assembly.

for out_prefix in `ls *.sort.bam | perl -pe "s/.sort.bam//"`; do mkdir -p ${out_prefix}-samtools-stats; for contig in `samtools view -H ${out_prefix}.sort.bam  | grep "^@SQ"  | cut -f 2 | perl -pe "s/SN\://"`; do echo "Now calculating stats for ${contig}/$out_prefix..."; samtools stats ${out_prefix}.sort.bam $contig > ${out_prefix}-samtools-stats/${contig}-samtools-stats.txt; done; done

• Pbiv1_RNase_WM_k141_107.fa

• Pbiv1_WM_k141_107.bam

• Pbiv1_WM_k141_107.bam.bai

• Pbiv2_WM_k141_23.fa

• Pbiv2_WM_k141_23.bam

• Pbiv2_WM_k141_23.bam.bai

---

Posts script- Vial 2 appears to be heteroplasmic for the 72bp insertion. Note. Vial 1 is depicted in the opposite orientation as Vial 1.

Jalview for viewing the annotated vector sequences.