Back around the year 2000,  there was this thing that was kind of a big deal in genetics:  the Human Genome Project.

Its goal was to decipher and publish the entire human genome for the first time ever. And people, including some  scientists and politicians, really hyped it up.

U. S.  President Bill Clinton claimed it would revolutionize the care of  almost every disease, for example. One kind of sky-high prediction  even said that, by 2016, we’d all carry around our own  personal genomes on a card in our wallets next to our driver’s  licenses and that doctors would be able to prescribe specific gene  therapies to anyone who needed them.

When we look back at those kinds  of news stories, it’s clear that the potential of the Human  Genome Project was overpromised. In this episode, we’ll talk about what the “Human Genome Project” was and why so much of the hype that was promised didn’t come to pass. But also, how it still kind of  revolutionized science anyway? [♪ INTRO] I don’t know about you, but I do  not have my genome in my wallet.

Though, to be fair, the Human Genome Project was, in many ways, a success. The project was an  international collaboration that ran from 1990 to 2003 to the  tune of 2.7 billion dollars. They did end up sequencing  the genome, or just about all of it anyways, which was  an incredible accomplishment.

But as the years rolled on, it became obvious that the promise to revolutionize how we approached every disease just hadn’t panned out. There was no magic bullet for  all our ills hidden in our genes. Why did some advocates promise this, then?

Some of it was no doubt hype or excitement. Even scientists and science  communicators can get caught up. But there was also some reason to buy in.

A couple of papers in the late  80s claimed to find single-gene causes for some pretty big medical issues. That included mental health  conditions like bipolar disorder, schizophrenia, and alcohol use disorder. If these conditions really did  come down to just a single gene, all we needed to do to treat them  was tweak that one gene, right?

Unfortunately, many of those  papers later came under scrutiny and their findings couldn’t be replicated. The truth is, most diseases are not  caused by a single mutated gene. A lot of diseases, especially  the big, common problems like diabetes or the aforementioned  mental health conditions, are influenced by genetics.

But they’re the result of  a lot of tiny nudges from many different genes, not just one big mutation. Today, we know that dozens  or even hundreds of genes may affect your risk of  developing diabetes, for instance. This means those genes do not lend themselves to a magic-bullet cure, because  just changing one of them probably won’t have any observable effect.

Worse, it may have unexpected side effects. Especially when you consider  that a person’s environment and life history also play a  role in their risk for diabetes. And scientists and doctors,  well, they kind of knew this was the case, even back in the 90s.

So you’ve probably noticed doctors  don’t routinely screen your DNA! Now, to be fair, there are some conditions, what we’d call Mendelian  conditions since they follow Gregor Mendel’s basic rules for inheritance, where a single mutation is enough to matter. Diseases like Huntington’s or sickle cell are caused by mutations in a single gene.

There are certain cancer genes, like the BRCAs, where certain mutations make  it much, much more likely for you to get breast cancer. And there are some medicines,  like the blood-thinner Warfarin, that work significantly better or  worse in people with certain mutations. Also, and this is cool, if you  already have certain cancers, doctors will sometimes do  genome sequencing on samples of the cancer cells themselves, since  screening for specific mutations can help your care team spot weak  points and choose treatments.

So there are times genome sequencing really does make a big difference in medicine. But it’s worth noting that, in all these cases, scientists already know  specific things to look out for. Doctors don’t really just blindly  sequence an entire person’s DNA and then go looking for oddities.

Because, to be honest,  knowing that a DNA mutation is there doesn’t really help until you  figure out what that mutation does. Like, let’s say you have a single nucleotide difference between two people’s DNA. Even if you know it’s part of a  gene – which we’ll come back to – it doesn’t necessarily tell you what the gene does or whether the mutation matters.

At this point, it’s all  still T-A-G-C gobbledy-gook. It’s possible that a scientist could look at that and make a guess about what that change would do. But unlike computer code, the genetic code doesn’t contain helpful annotations  to say “this part does this.” You can’t tell what the genome does by reading it.

You have to sit down and do experiments to see what a change functionally does. It’s much easier to start with a  known disease – like diabetes – and then work backwards, looking  for mutations in proteins we already know are important  or comparing and contrasting a lot of sequences from people  with and without the disease. So that’s what we mean when we say people back around the year 2000 probably overpromised things about the Human Genome Project.

The project did successfully sequence the genome, but saying it was going to  revolutionize how we treat nearly every disease was skipping  a lot of steps in the middle. And there was something else they couldn’t really have known just yet. Brilliant and SciShow have a lot in common.

We’re both excited about  fun and accessible science. And we both prioritize accuracy in  our online learning opportunities. We also like to branch out and  try learning about new fields.

Like when we made the SciShow  Psych and Space channels. And Brilliant just came out with new courses on data, programming, and large language models. The same way you could spend hours going through SciShow’s back catalog, you can spend hours on Brilliant’s  online learning platform, going through their thousands of lessons in science, math, and computer science.

So for pretty much all of the  reasons that you like SciShow, you’ll probably like Brilliant too. You can check them out at Brilliant.org/SciShow or the link in the description down below. That link also gives you 20% off an annual premium Brilliant subscription.

And you’ll get your first 30 days for free! Now back to the show! Here’s part two of the video,  because what scientists didn’t know was what all those  letters in the genome actually do.

Now stop me if you’ve heard this one before, but it’s a really common metaphor to refer to DNA as a kind of writing and the genome as a book. Metaphors are good! They’re a good starting point  our brains can hold onto as we learn something new and complicated.

But metaphors are also at  least a little bit fiction. And it turns out the real genome  is much weirder than a book. The human genome is about 3 billion letters long.

At the time, it was assumed that most of those letters would code for proteins. As a quick reminder, DNA gets read out or transcribed into messenger RNA, which gets processed and then further read out or translated into protein, and proteins are what do much of the work of being alive. So it was not actually a stupid  assumption to think most of the genome would code for protein.

But, hoo boy, it does not. The sequence of the genome revealed that the vast majority of our DNA –  something like 98 or 99 percent – doesn’t code for proteins. It was as if most of the pages of  the book were filled with nonsense, what some scientists called “dark  matter” or, more pejoratively, “junk”.

Firstly, for instance, there’s  DNA sequences that affect how genes are expressed without  actually being genes themselves. Promoters, for instance, are stretches of DNA just ahead of the protein-coding part of a gene. They are where a bunch of cellular  machinery will come sit down to read that gene out, a little  like markings on a runway.

There are also enhancers,  repressors, and silencers, which are sequences of DNA that either increase or decrease expression of a gene. There are also parts of the DNA  sequence that are transcribed into mRNA, but, for some reason,  do not end up in the protein. Introns are DNA sequences that get cut out of the mRNA before it gets translated into a protein.

Sounds weird, but it’s actually a flexible system that can lead to different  versions of the same protein by chopping the RNA up in different ways. Sometimes DNA codes for RNA that has a specific job other than becoming a protein. For instance, the ribosomes that  translate mRNA into protein are, ironically, themselves mostly made out of RNA!

There are also bits of DNA  that seem to be more like cellular scaffolding or  structure than instructions. Like telomeres, which are long  sequences of repetitive DNA that kind of cap off the end of a chromosome  and keep it from unravelling. And, finally, yes, some of it  does seem to actually be junk.

There are things known as pseudogenes, which seems to essentially be “broken”  genes that no longer function. Even the 3D architecture of the cell’s nucleus can affect gene expression. Like very long noodles in the  worst plate of spaghetti ever, chromosomes have to be arranged  in 3D space in order to work.

The enhancers I mentioned earlier loop around from really far away to affect  the genes they regulate. In contrast, DNA that’s packed  up really tightly is harder for cellular machinery to get  at, so it’s expressed less. And all that’s just scratching the surface.

There’s more weird stuff  going on, as well as stuff we still don’t quite understand yet. But the point is that the  genome is really complicated, and it’s about a lot more than  reading letters from beginning to end. Basically, our book is making House  of Leaves look straightforward.

So here’s an alternate metaphor.  It’s still a bit of a fiction, because all metaphors are a  bit of fiction, but humor me. The genome may be less like a  book and more like a library. There are books with blueprints, and people copying down what they see inside.

But there is also scaffolding and shelving. There’s Dewey Decimal Numbers and reference maps, locked-up sections and big  open promotional display cases, and internal documents and memos. You might be able to see a few  librarians reshelving things or shuffling them around  and, yes, even a few tattered “junk” books destined for recycling.

I think that’s a much cooler concept than a book. So combining both the points about diseases and the complexity of our DNA, we arrive here. In sum, knowing the entire  sequence of a patient’s genome would likely do very little for their  care outside of certain situations – and that probably won’t ever  change much in the future.

The big diseases are usually the  result of many small nudges, rather than one big push, and not all of  those nudges lie in the genome. And the genome itself is really  complex, so even if you know of a mutation, that doesn’t  necessarily tell you that much. So that specific promise, about  being able to sequence everyone’s DNA leading to easy cures to  everything, never came to pass.

But… here comes the twist. Does that mean that the HGP a failure for science? No!

It does not! All that stuff I just talked about, we were able to learn about  because the Human Genome Project provided a scaffold for  everyone else to work from! In many ways, the Human Genome  Project, like the moon landing, was primarily a technical achievement.

The techniques they used paved the way for modern sequencing technologies. Today, what once cost billions of  dollars and took two years is down to less than a thousand dollars and  can take as little as five hours. A scientist can whack a sample  down in a fancy sequencer, go have a sandwich and look  after a few other experiments, and come back to results before  it’s time to quit for the day.

And this proliferation of  cheap, reliable, and fast genome sequencing did indeed  revolutionize science. The tools we got out of it are used  in everything from understanding basic cellular processes to studying  evolution to ancient archaeology. What’s more, going wide and  being able to test the genomes of a lot of different people  may deliver some of the benefits that we didn’t get from just doing it once, because there’s power in being  able to sequence a lot of people.

For one thing, it helped us better grasp the diversity of the human gene pool. The original sequence the  Human Genome Project published was a patchwork of just a handful of people – actually, about 70% of the genome came from one anonymous sample. Today, though, subsequent efforts  like the 1000 Genomes Project give us a much broader sample size to work with, allowing us to understand not  just one person’s genome works, but how genetics affects  people all around the world.

The Human Genome Project era  also gave way to the era of what’s known as genome-wide  association studies, or GWAS. In general, this involves  rounding up a bunch of genomes and looking for patterns or associations between a disease and a bunch of genetic variants. Results from GWAS may translate  into discoveries for diabetes, autoimmune disorders, and schizophrenia.

Like, that’s how we got those  hundreds of diabetes genes. The contribution of each one  is so small that a pattern only emerges when you’re looking  at loads and loads of people. There still will probably never  be a single magic bullet gene therapy-type thing for every  disease, but these discoveries can give us a better understanding of how these diseases occur and the best ways to treat them.

The Human Genome Project  helped scientists at the lab workbench more than it helped  patients in the doctor’s office. Of course, the scientists at the lab bench do help the people in the doctor’s office eventually. It’s just more roundabout than we’d hoped.

So that’s the story. The Human Genome Project was a huge achievement that  may have been over-hyped, but which still was a huge technical success and is still leading to exciting new things. That said, with the benefit of hindsight, we’ll resist making any huge promises about what’ll end up in your wallet in the future.

As we’ve learned, genetics can be far more complex than it initially appears. If you liked this episode about  why the Human Genome Project was a failure but still a good  idea, you might also like our video about why the moon landings were  a failure but also a good idea. Thanks for watching. [♪ OUTRO]