Most vertebrates have hundreds of endogenous retroviruses in their genomes. Almost none of the endogenous retroviruses are fully functional and able to produce infectious viral particles, but many of them have potentially functional reverse transcriptase genes. Despite this, the reverse transcription of messenger RNA to DNA, and integration of that DNA into the host chromosomes, is an extremely rare event. Vertebrate genomes do contain some pseudogenes which are reverse transcribed copies of messenger RNA, but not a lot of them considering the hundreds of millions of years these genomes have been evolving.

There is more to reverse transcription and integration of DNA into genomes, than just having an RNA and a reverse transcriptase enzyme in a cell. For one thing there needs to be a primer to start the reverse transcription process. HIV-1 for example uses the host cell lysine transfer RNA as a primer and HIV-1 has a region that is complementary to the lysine tRNA to allow this to happen.

RNA:DNA heteroduplexes which are the intermediate step in creating dsDNA from RNA during reverse transcription, are not typical in the eukaryotic cytoplasm. They are recognized as foreign and destroyed by the host cell innate immune responses. Viruses have to evolve mechanisms to defeat this process in order to be successful. HIV-1 for example has genes such as the vif gene which defeats the host cell APOBEC system.

One caveat to this is that we generally only have sequence data for events that have been fixed in the germline of our reference assemblies.

We don't necessarily know what the rates of these "rare" events are in somatic cells of various lineages.

We would not have observed these because we aren't generally looking for them.

It's also the case that reference-based-resequencing (ie: current NGS short-read sequencing) would result in these things being thrown away as artifacts. This is particularly true of chimeric/multi-mapped reads. Also possible that it could result in an excision event which might be difficult to recognize.

So I agree with the characterization of these as rare, but it often turns out we've underestimated the probability of these sorts of rare events, and have been supported in doing so by biased data.

I'm having double thoughts on HIV patients taking sarcov2 mRNA vaccine, could there be any cascade of reactions between the two viruses? Just thinking thought!

I do not have an answer to the question but the questions seems to me very relevant since the main idea is that mRNA vaccins will never be able to integrate in host DNA. The question that I derive from your question is whether this is always true, So also when we take into acount the diversity of retroviruses which may have infected a part of people that are unaware of this. Is there any possibility that these retroviruses may interact with the mRNA in the vaccin, and by this built in the transcription of the mRNA vaccin into the host DNA. Next question would than be, whether this may have long term or even generational health consequences.

Victor van Rij , all of our cells are making messages (and other RNAs such as ribosomal RNAs) all of the time. If reverse transcription of RNA and its integration into cellular DNA was a problem, our genomes would degrade in a matter of days and life would not go on.

There is more to the story than just reverse transcription, there is also the creation of the second strand to make double-stranded DNA (an RNA-DNA heteroduplex is not going to work), importation of the molecule into the nucleus, and integration.

The process by which retroviral RNA creates a double stranded-DNA circular molecule and uses its virus-encoded integrase enzyme to insert itself into host cell chromosomes is complex. One of the very first key elements is that the retroviral genome contains a site complementary to a host cell RNA (HIV-1 for example has a region complementary to the 3' end primate Lys-3 transfer RNA) so that this host cell RNA can serve as a primer for the reverse transcription process.

HIV-1 also encodes a Vif protein which is critical for evading the host APOBEC system which detects and destroys RNA-DNA heteroduplexes found in the cytoplasm.

Anyway, our cells contain thousands of RNAs all the time, so one more RNA is not "especially dangerous" for any reason. And also, consider what happens when a real virus infects us. The real virus not only puts RNA into our cells, but that RNA replicates and kills our cells, it spreads to other cells, etc... The vaccine is thus in theory about a million times safer than allowing a real virus to attack us.

Brian Thomas Foley , I am not doubting about the fact that the vaccin is much safer than a real infection. Of course we are talking about very rare processes, that are mainly of interest from the evolutionay point of view. I already was thinking, that if there is any chance that the mRNA string (used for the vaccin) would cause any change in the genoom of the host, this would be certainly the case if someone would get a coinfection of the whole coronavirus and the retrovirus. The same would apply for all other coinfections of RNA viruses. Of course it would be very rare but from evolutionary perspective interesting. It would be interesting to search for matches in a population of host genomes with parts of a variety of RNA viral sequences to see if despite the complex specific interactions you mention, "rare" integration of DNA copies of RNA viruses or mRNA sequences of viruses in host genomes have occured. Of course it will be hard to see if this has any impact on a individual scale. I think that anyhow it is a reassuring thought, that the vaccin long term effects (if any) of the mRNA Covid19 vaccin wil be not be any worse than the longterm effects that "may" come from a full infection.

If HIV RT create antisense DNA and it gets integrated in area that allow further expression. Subject might continue to produce spike protein. Knowing that it is membrane anchored, it is likely that immune system will delete these cells, unless in the process the cDNA lost the trans-membrane domain and this will be a problem.

Victor van Rij, there have been many studies of mammalian genomes (all of which contain many endogenous retroviruses with potentially functional reverse transcriptases and integrases), searching for integrated viral genes or genomes.

https://pubmed.ncbi.nlm.nih.gov/27217174/

https://pubmed.ncbi.nlm.nih.gov/25675979/

Article Paleovirology - Ghosts and gifts of viruses past

Article Virus-like insertions with sequence signatures similar to th...

Article Virus et gènes viraux endogènes: Évolution et impact

etc...