RNA

because I think it's remarkable because it spans the typical academia lifestyle for a long while, going through industry, entrepreneurship, and now also scientific publishing, editorial work.

Going back in time, what drew you first to science and RNA biology or RNA therapeutics?

Yeah, yeah.

And thank you, Marcin, for having me here.

So you've known me for probably close to 15, 20 years now, which dates both of us.

But anyway, yeah.

So how did I get into science and ultimately RNA biology? Ever since I was a little girl, I loved nature, animals, trying to understand how things worked.

The human body fascinated me.

Okay.

So the test basically would extract all the genetic material, so all the RNAs, the DNAs of the organisms, and We'll link them to one organism.

Okay, so this is a lactobacillus.

This is a streptococcus.

Now, the way that that works, understand an interactome, where The genome and the epigenome, if you look at them as on two sides of a horizontal axis, they're always interacting.

You know about the epigenome, covalent modification of such things as cytosine residues on DNA, adenosine residues on DNA and RNA.

And of course, various amino acids, particularly we talk about lysine, but of course, serine and arginine are involved as well as other amino acids like threonine on histone proteins as well as transcription factors.

And all the other proteins that are involved in the effects associated with DNA biochemistry, DNA replication, DNA recombination, DNA repair, and the transcription of DNA to RNA.

Remember all those epigenomic alterations we spent some time on the last couple of lectures.

So that's the genome-epigenome interaction.

Why am I saying both epi and just plain on all those? It's because, remember, the genome is dictating, but the epigenome is coordinating that written formula.

What's the written formula? The chromatin.

These days, we have big patient data sets like TCGA, the Cancer Genome Atlas, that we can use that to start.

So what are the genes that are overexpressed in cancer compared to the corresponding normal tissue? That's just one thing, right? That's RNA.

But then could you look at the RNA expression profiles of other normal tissues and not just the paired tissue of origin? So if it's liver cancer, hepatocellular carcinoma compared to non-tumor-bearing liver, right? But what about other tissues, right? Because you might care about, you know, if you have expression in your salary glands, for example, you might want to know about that and you take that into account when you're choosing your target.

Got it.

So that's the target.

Why is this is not talked about? Is it because that repeated vaccination has caused stimulation of the T suppressor, now called T regulatory cells, that down regulates the immune response? Is that what is happening, this opposite effect of vaccination? Or is it that repeated vaccinations are stimulating this rather new antibody we're learning about, immunoglobulin, G, immunoglobulin G type 4.

Now this, I've only heard about this in the context of RNA viruses, RNA vaccines, messenger RNA vaccines.

It didn't seem to arise after the adenovirus vector vaccines.

So it's maybe something specific.

It could well be another weakness, of course, in this whole whole trend to go towards mRNA vaccines.

But how is this working? Well,

if...

Infected cells are less likely to be eradicated.

Now, it's a shared facility and peer-reviewed in nature.

Add one more piece, an RNA model in nature called StructRFM.

It trained on 21 million pairings of RNA sequences with their folded shapes because a molecule shape affects what it does.

A 3D predictor it supports performs comparably to AlphaFold3 on three standard benchmarks.

Comparable, though, not quite better.

Twice yearly HIV prevention at about 40 bucks a year.

A lab many people can share.

RNA models anyone can download.

But I was not enjoying taking care of all the flies, because you can't just freeze flies away, thaw them out, and use them again.

I was just working with RNA and DNA, and I wanted to do something behavioral, more with mammals.

So I started looking around for a postdoctoral position in some kind of mammalian behavioral biology to get away from flies.

I had looked at a few different opportunities, and just through a series of serendipitous events, ended up in a lab at Vanderbilt, Elaine Sanders Bush, who had been studying psychedelics since the 1960s.

So when I say unsupervised data, this is basically like People have, you know, scientists have done the heroic work of curating all of this basically DNA data that you can then turn into proteins because that's what biology does.

It takes DNA, turns it into RNA, turns it into proteins.

And so you can say, OK, if I see this DNA, then I know this protein was out there.

um um and then they've curated it all in in these massive databases where you have you know uh 250 million uh different protein sequences that that are have been observed out in the world Um, and so you can take all of that unsupervised data and you can train a model that kind of knows what plausibility is like.