Here Is a Human Being (37 page)

When I got my complete genome from the Goldstein lab, one of the first things I did was to check my genotypes in the RET gene, the major Hirschsprung’s susceptibility gene I had worked on. They were completely normal. And even if they hadn’t been, what conclusion could I have drawn, other than that God had a twisted sense of humor? Jesse Angrist was not Bea Rienhoff. He did not need his father’s—let alone his uncle’s—heroics. Most of us were fortunate that we didn’t.

The other “rare” change turned up by the PGP that was connected to disease was in one of the dozen or so genes responsible for Fanconi anemia, an autosomal recessive disorder characterized by short stature, skeletal defects, a higher incidence of cancer, bone marrow failure, and cellular sensitivity to chemicals known to damage DNA.
⁷
Jim Watson, Rosalynn Gill-Garrison, and the first Asian genome (Chinese) carried it as well. Since it was recessive and I was done having kids, it was not a major cause for concern (I don’t have Fanconi anemia, although I am short). Of course, it was entirely possible that one or both of my daughters carried it; I considered that to be something worth remembering.

And what about my own kids? I had gotten good news from the breast cancer mutation test: I did not carry any of the three most common mutations found in Ashkenazim. But BRCA1 and 2 were big genes. More than 1,600 distinct variants had been reported in BRCA1
⁸
and more than 1,800 had been reported in BRCA2.
⁹
In each case more than half of them had been seen only once. Rare was usually good when it came to cancer, but of course rare was also what so often went undetected. I was a bit taken aback to see a fairly rare variant in one copy of my BRCA1 gene. At amino acid position 1564, I appeared to carry a change that altered a histidine to a proline, a “nonconservative” change from a basic amino acid to a neutral one. I became a bit agitated: After my DNA Direct test for the Ashkenazi mutations, I had more or less dispensed with worrying about being a carrier of a breast cancer mutation. Had I exhaled too soon? Right away I began digging into the breast cancer genetics literature but couldn’t find much. In part this was because what the Trait-o-matic had shown me included a typo: the actual mutation led to a leucine-to-proline change in the BRCA1 protein, not a histidine-to-proline one (not that that was necessarily any better or worse). Once I figured that out, I went looking for papers that had analyzed L1564P instead of H1564P. In a study of African American women, L1564P had been seen in patients with breast cancer but not controls.
¹⁰
This was interesting, but not authoritative: the sample was small and the mutation detection method used in that study was not perfect. There was nothing to say this wasn’t an innocuous variant. Or a death sentence. Heh. “Absence of evidence is not evidence of absence.”

I needed a higher power and I found one: the Breast Cancer Information Core (BIC), a consortium hosted by NHGRI that had taken upon itself the job of determining the clinical significance of every single one of the thousands of variants observed in the breast cancer susceptibility genes. I filled out the online application, gave my Duke credentials/affiliations, and assured BIC that I was only doing “research.” Once inside I easily called up L1564P and found a wealth of information about it: how many times it had been seen (twelve), in what populations (mostly African), what did this type of change mean for the BRCA1 protein (most likely nothing), and the bottom line: “Overwhelming evidence from sequence conservation, epidemiologic studies, [and] co-occurrence with different deleterious mutations [suggests] that this variant is not a significant cancer susceptibility allele.” Woo hoo!

BIC turned out to be a wonderful resource. It epitomized what a genomic database should be: it housed all of the necessary and sometimes inscrutable reams of population genetic data, but it also showed what could be done when information was not only aggregated but interpreted with an eye toward actual human beings’ health. Of course, BIC cautioned against using its information for clinical purposes and despite being labeled “open-access,” BIC was available only to “qualified investigators.” We can’t have civilians poking around in these minefields, now can we?

In my whole-genome data, I found another change in the protein-coding portion of BRCA2 and emailed breast cancer geneticist extraordinaire Mary-Claire King again. She didn’t remember me and probably thought I was yet another kook bugging her for wisdom (which I was, but never mind). She kindly informed me that the variant I carried was seen with equal frequency in both breast cancer cases and controls, that is, almost certainly nothing to worry about.
¹¹

But mutations don’t happen only in protein coding regions. If they occur in introns (the discarded parts), they can still mess up the transcription process that gets DNA to RNA. And so I went exon by exon (BRCA1 has 24 exons; BRCA2 has 27) and intron by intron and took a look at each of the eighty-six SNPs and twenty-two indels I carried. I could ignore many out of hand because they had already been seen in other databases full of healthy people. As for the rest of the variants, the sequence and SNP databases—even BIC—often weren’t much help. The clinical significance of these changes, their frequency in the population, and even whether they were passed down in predictable Mendelian fashion—all of these facts were often unknown. It turned out to be a couple of hours of work that was somewhat reassuring—the chance I carried a “non-Ashkenazi” mutation was pretty small to begin with. But even this was not an unconditional guarantee. And frankly, even for a genome geek the whole business got to be tedious. I was ready for the dial-up age to be over.

Thanks to Dongliang’s patience, I was finally able to download selected other subsets of data. One was a file of all of the variants in protein-coding genes that inserted or deleted bases and disrupted the proper reading of amino acids—so-called frameshift mutations. In 148 instances I carried two copies—one from each parent—of these types of mutations. In twenty-seven of those, the frameshift introduced a premature stop codon as it did in the Factor VIII example Dongliang and I had looked at; presumably premature-stop mutations would yield no protein. Francis Collins had warned me a year earlier: “You will have some
breathtaking
mutations.” The thinking was that if one had zero functional copies of something instead of one or two, then that might be expected to have an effect. Or not: I found that I carried two defective copies of the CASP12 gene, which is an important gene involved in inflammation. But there’s a good chance that you do, too. Most humans carry mutated versions of CASP12 and produce a truncated, nonfunctional protein. The genome often makes do with less. And perhaps it’s not just making do: nonfunctional versions of genes like CASP12 may actually be a selective advantage. This might help to explain why each of us walks around with dozens and dozens of broken genes.
¹²

I kept on in hand-to-hand-combat mode, going gene by gene. And time and again, the paucity of information was striking: I would find mutations in genes that coded for proteins, but the proteins’ ascribed functions would be so general and/or tentative (“may be involved in transcriptional regulation”) as to be meaningless. In some cases, the proteins didn’t even have names, let alone functions assigned to them.

Another striking realization was that our genomes are an utter mess: chunks of DNA are not only missing, but they have often moved to different chromosomes, or flipped around. Genes we don’t need—ones that code for olfactory receptors, for example—may be completely absent. Why don’t we need them? Because we have hundreds more where those came from, for one thing. And in evolutionary terms, other than for detection of the ocasional noxious fume, we don’t necessarily need a sense of smell to survive anymore. The aroma of food can be pleasurable (pizza, popcorn, fresh-baked bread, curry) or repugnant (rotten fish, rancid Limburger cheese, your brother after he’s eaten a burrito), but these days Westerners generally don’t depend on it for hunting and gathering.

The genome was also a source of pure banality. We all have “housekeeping genes” that are turned on all the time at the same level, doing the same boring job in nearly every cell in nearly every tissue. Like olfactory receptors, these genes are often redundant; one fewer copy would probably not be missed.

And some genes appear to be present but not real: they are pseudo-genes. They typically have many if not all of the hallmarks of a gene: a beginning, middle, and end; a plausible sequence of amino acids; a regulatory region; binding sites for proteins. But these “genes” don’t lead to RNA and therefore they don’t lead to protein. There may be every reason to believe that the proteins encoded by them should exist, but no one has ever seen them.

This often-maddening state of affairs triggered an analogy that kept popping into my Excel-weary brain. On a windy summer day near Harvard Square, I met Andrea Loehr, a blond German who was volunteering as a data cruncher for the PGP and who had been to the South Pole several times. She was trained as a cosmologist and explained that Antarctica affords astronomers a view that can’t be had anywhere else on earth: if one were interested in looking into space, she said, the South Pole was a great place. “It’s high, it’s dry, and the atmosphere is extremely stable.” The problem—and the reason why she decided to switch from stargazing to whole-genome sequence data analysis—is that in cosmology, getting meaningful answers can take decades. “In fact, I may not live to see them,” Andrea said. “I like the idea of a project that will have at least some return within a few years.”
¹³

A few months earlier Hugh Rienhoff had brought up Antarctica to me for a different reason. He had been doing some consulting for Knome and had seen a few of the company’s first customers, the ones who had shelled out $350,000 to get their complete genomes done. “What motivated them?” I wondered. “It’s curiosity,” he said. “People want to go to Antarctica. Why? They like to do it. It’s icy, it’s cold, it’s windy, it’s dangerous. People getting their genomes sequenced is that kind of thing. Though I don’t think they’re learning much they didn’t know already.”
¹⁴
Or perhaps they were.

When most people think of Henry Louis “Skip” Gates, Jr., chances are they think of his arrest for disorderly conduct in the summer of 2009, the surrounding kerfuffle, and his subsequent beer with the arresting officer and President Obama.
¹⁵
They may know that he is the Alphonse Fletcher University Professor and the director of the W. E. B. Du Bois Institute for African and African American Research at Harvard University. That he is a dignified and frequently feted scholar, the winner of a MacArthur Genius Grant.
¹⁶
They might also think of his
African American Lives
and
Faces of America
documentary series,
¹⁷
which seem to commandeer PBS every other February (certainly more fun than pledge drives), and the famous Americans who participate in them.

What they may not realize is that Skip Gates is hilarious and a keen student of genetics.

As we chatted in the kitchen of his immaculate (and, in the wake of death threats, now quite secure) yellow house, a stone’s throw from Harvard’s sprawling and semistately campus, Gates made himself a high-fiber, low-fat tuna salad sandwich while I sat at the far end of the large white island in the middle of the room. Soon the doorbell rang. He looked at me.

“If it’s the police, tell ‘em I ain’t here,” he deadpanned.
¹⁸

Gates came to see molecular genetics as a tool to uncover African American ancestry in 2000, when he connected with Rick Kittles, then a geneticist at Howard University just beginning molecular studies of African ancestry.
¹⁹
But Gates’s interest really began at age nine, “the day my grandfather was buried.” The next day he began work on the Gates family tree. When Alex Haley’s
Roots
was broadcast in 1977, Gates was riveted. As he got older, he became a devoted Africanist, eventually visiting twenty-one countries on the continent. When Kittles explained what he was doing, “I was

down,”
said Gates. “I said, ‘You’re gonna take a little blood and tell me what tribe I’m from?’ That’s what I’m talkin’ about.” He soon donated blood.
²⁰

Time went by and Gates called Kittles. “Hey, man. Where’s my Kunta Kinte moment?”
²¹
In 2000 the database of genetic markers was a shadow of what it is today. And in the relatively early research phase of his studies, Kittles looked only at markers common among Africans. Consequently he could not identify a strong African link to Gates’s maternal line.
²²

“I thought it was a simple test: you take some spit and a tribe lights up in Africa,” Gates recalled. “I didn’t know about private and public genetic databases. It wasn’t my field. But some of the time I’m smart enough to know what I don’t know. I realized it was time for me to pull back, not treat the subject cavalierly, study the science myself, and surround myself with a battery of experts, some of whom disagreed with Rick.”
²³

Subsequent analyses all indicated that Gates’s roots were in Europe, and the British Isles in particular. Enthralled with the science of heredity, Gates took the opportunity to school himself. He took the PGP exam and a DVD-based course in undergraduate genetics. He asked the Broad Institute’s David Altshuler, Mark Daly, and Eric Lander to mentor him.
²⁴
And George Church, too. “He had an endless series of questions and at times it seemed like it was an infinite loop,” recalled George with a smile. “Haploid this and diploid that, ‘Were genes like twenty thousand volumes in a library?’ and ten other analogies going simultaneously. He seems to have only one setting on his potentiometer, which is
ecstatic.
I think it’s going to be a good thing to have him involved.”
²⁵