I walked into a conference room at Yale University not long ago to find eight graduate students and postdoctoral researchers waiting for me on either side of a long table. They invited me to sit at the head. In front of me, on the opposite wall, was a giant monitor. On it read the words, “Individual Z Overview.’’
For two weeks, these researchers had been poring over my genome, and now they were ready to share with me what they had found. I had been gratified by how eager they had been to help me, but puzzled, too. It was only when I looked up at the screen that I realized the answer. To them, I was Individual Z.
It was as if I were a frog that had hopped into an anatomy class with my own dissecting scalpel, asking the students to take a look inside.
The students all worked at a lab run by Mark Gerstein at Yale. I had recently had my genome sequenced and asked Gerstein to look at it because he has studied thousands of human genomes over his career.
He and his colleagues are experts at making catalogs of genomes — recognizing the genes and millions of other pieces that make them up, and figuring out how those parts vary from one person to the next. Yet when I approached Gerstein about my project, he admitted he had never looked at his own genome this way.
“I’d never have the courage to do this — I’m just too timid,’’ Gerstein admitted to me. “I’m a worrier. Every time there would be a new finding, I’d look in my genome to see if I had it.’’
While Gerstein might be too nervous to look at his own genome, he seemed to take vicarious pleasure in looking at mine. “I really want to do this,’’ he said when I handed him the hard drive with my genome’s raw data. “I think this is the future.’’
In getting my genome sequenced, I had managed to obtain what’s known as a BAM file — an enormous file containing all the raw data of my genome.
Gerstein transferred the data onto his computer and gave the hard drive back to me. He and his team then used a set of computer programs to analyze my BAM file and build a highly accurate reconstruction of my genome. Once they had reconstructed the sequence of my DNA, they could start identifying the parts that made it up.
The parts of a genome we are most familiar with are, of course, genes. Each protein made by our bodies — such as the collagen in our skin and the myosin in our muscles — is encoded by a gene. Our 20,000 or so protein-coding genes take up only about 1 percent of our genome, however. They are scattered amid vast stretches of so-called noncoding DNA. Noncoding DNA is a mishmash of different elements. Some of them, like on-off switches for genes, are essential to our well-being. A lot of them are just along for the ride.
In order to map the parts of my genome, Gerstein and his colleagues took advantage of the fact that one person’s genome is pretty similar to anyone else’s. If you want to find your COL1A1 gene for collagen, for example, you’d best look about midway down your chromosome 17. That’s where it is in everyone else.
But while my genome is a lot like everyone else’s, it’s not identical. When a scientist like Gerstein sets out to catalog a genome, a lot of his work goes into tallying up my differences.
When I returned to Gerstein’s lab for my Individual Z Overview, Fabio Navarro, a Brazilian postdoctoral researcher with a scruffy beard, kicked things off by introducing me to a big number: 3,559,137. That is how many positions in my genome differ by a single base from the human reference genome — a single nucleotide polymorphism, or SNP for short.
For example, I have rare SNPs in a gene called MEFV. At one location in that gene, the vast majority of people have a base called thymine. But one of my copies of the MEFV gene has a cytosine at that spot. This variant gives me the rare distinction of being a carrier for a disease called familial Mediterranean fever, which causes runaway inflammation. (You need two copies to actually get the disease.)
It was a struggle for me to think clearly about the 3,559,136 other SNPs in my genome. I was tempted to think of them as making me an exquisitely unique genetic snowflake.
Sushant Kumar, another postdoctoral researcher who works with Gerstein, dispelled that illusion by picking out two people from a database of genomes to compare me with. One was a person from China, the other from Nigeria. Kumar found all three of us shared a lot of SNPs in common — 1.4 million, in fact. Kumar and his colleagues cut down my uniqueness even more by searching for my SNPs in a database they helped build, called the 1000 Genomes Project. They found more than 91 percent of my SNPs in at least one other person’s DNA.
This enormous genetic overlap is the result of humanity’s sloshing global gene pool. Every new baby gains a few dozen new SNPs. They can pass on some of those SNPs to their own children. Over thousands of years, the variants spread from continent to continent.
Gerstein and other scientists want to understand how these SNPs influence our body. For now, most of what we know about the variations is limited to our protein-coding genes. But even in that 1 percent of our genome, our knowledge is limited.
I discovered, for instance, that I have a variant in a gene called HMGA2 that makes me a little taller. On average, people with my variant are about a quarter of an inch taller than people without it. But scientists don’t yet know exactly how it boosts the growth of people like me who carry it.
It’s likely that variants like the one in my HMGA2 gene influence my biology by changing the shape of my proteins. When proteins change shape, they work differently. During my visit with Gerstein’s team, one of his graduate students, Declan Clarke, provided me with a startling demonstration: He showed me the shape of some of my mutant proteins.
One of my mutations changes the shape of an enzyme in my liver. Our livers keep our blood clean by breaking down potentially harmful molecules so that they can get flushed out of our bodies. One of those enzymes, called NAT2, helps break down caffeine and other toxins with a similar molecular structure.
I have a variant in my gene for NAT2 that changes the enzyme’s shape. Clarke showed me how a pocket on my enzyme has an odd bulge. That bulge changes the way my NAT2 enzymes behave. In most other people, that pocket repels water molecules. In mine, it attracts them.
As a result, my NAT2 enzymes work slowly, allowing toxins to build up and linger longer in my body. Making matters worse, my defective pocket raises the risk that NAT2 enzymes will stick to each other, or to other proteins. To protect me from this damage, my cells destroy a lot of my NAT2 enzymes.
“Let’s say you have an old beat up car and you’re driving it around on the road,’’ Clarke said. “It’s like the other proteins are saying, ‘We have to impound this thing.’ ’’
While getting rid of a lot of my NAT2 enzymes might reduce my risk of dangerous clumping, it also leaves me with even fewer of them. As a result, I end up doing an even worse job at breaking down certain toxins. And it’s not just toxins that can pose a problem: NAT2 helps break down certain medicines, too. Geneticists have found that my variant puts people at risk of bad side effects from those drugs.
While some mutations alter proteins, others destroy them. They disrupt genes so badly that our cells can’t use them to make any functional proteins at all.
A broken gene (technically known as a loss-of-function variant) can be a very dangerous thing. If you don’t have a functional F8 gene, for example, you can’t make an essential clotting protein. You get hemophilia and can bleed to death from a little cut.
In my own genome, Gerstein and his colleagues discovered 13 genes in which both copies appear to be broken. I have another 42 genes in which only one copy looks like it’s defunct.
It might sound strange that my genome has dozens of broken genes that cause me no apparent harm. If it’s any consolation, I’m no freak. The 1000 Genomes Project revealed that everyone has a few dozen broken genes.
Our genomes are not finely engineered machines that can’t tolerate a single broken flywheel or gear shaft. They’re sloppy products of evolution that usually manage to work pretty well despite being riddled with mutations.
I’ve probably passed down some of my uniquely broken genes to my children. Perhaps, long in the future, one of those broken genes will become more common in humans and end up in every member of our species. That’s certainly happened in the past. My genome catalog includes about 14,000 genes that have been broken for thousands or millions of years, known as pseudogenes. Once they lost the ability to make proteins, they simply became extra baggage carried down from one generation to the next. Thanks to a genetic roll of the dice, they ended up becoming common. Now these 14,000 pseudogenes are found in all humans today.
“It’s neat — this is evolution in process,’’ Gerstein said. The unbroken continuum from my own broken genes to humanity’s shared pseudogenes is testament to the long, error-filled journey that produced our complicated, baffling genomes today.
By Carl Zimmer | STAT
About this
series
In Game of Genomes,
STAT national
correspondent
Carl Zimmer takes
a narrative journey
through the
human genome —
his own. The first
journalist known
to have acquired
the raw data of his
own genome, Carl
spent months
interviewing
leading scientists
about the latest in
genome research
to learn more
about himself and
about human
genomes in
general. This is an excerpt from the second part of the three-part series.
To read more, go to www.statnews. com/gameof genomes.
Carl Zimmer can be reached at carl.zimmer@statnews.com. Follow Carl on Twitter @carlzimmer.
