Adventures in Recreational Genomics

Part 3: Pandora's Box

Add to Shopping Cart

There is no stopping scientific advances in our search for the meaning of our genetic code. This is the frontier, and the FDA, medical establishments, and consumers may just not be prepared. And with the costs declining rapidly, the information is now more widely accessible.                  

Declining cost per single genome sequence

Over a decade ago, Dr. Craig Venter with his company, Celera, raced against the National Institutes of Health to sequence part of the human genome. It cost the NIH about $3 billion in taxpayer money, and Venter about $300 million from private funding. Costs have plunged and access is increasing since. In 2011, Steve Jobs had his entire genome sequenced for $100,000 in an effort to treat his pancreatic cancer. You can now submit a blood sample to Gene by Gene, Ltd., of Houston, and pay $7,395 to have your entire genome sequenced. (I love the “Add to Cart” button on their site where you purchase the service.)

Public and private companies are competing to develop full genome sequencing platforms that can be used for both research and clinical use at an acceptable price point, these companies are heavily backed by venture capitalists, hedge funds, and investment banks. In 2014, the $1,000 cost barrier was broken. This price tag was seen as vital to making whole-genome sequencing cost-effective for widespread medical testing and personalized medicine.

It was San Diego’s Illumina, Inc., the company that provides the simple chips 23andMe uses for SNP analysis that broke the barrier. They launched the HiSeq X Ten Sequencer, which delivers one complete genome, 20,500 genes, for about one grand, which is basically the cost of chemicals and sample preparation. It has much better optics and faster chemistry than other systems, and can analyze three complete human genomes every 16 days. After Illumina announced this new product, the company’s share price jumped to an all-time high of $182.

Think you might want to buy a system? You have to purchase ten machines at a time, at $10 million for the package, and then a million bucks for each machine after that. No discount. Of course, you have to store the data, and that’s estimated at about $60,000 for server space for one month’s output. If I’m doing the math correctly, that’s only about six peoples’ genomes worth of A, G, C and T’s on that one $60,000 server. And you’ll need lot more cash to pay the techies and scientists to interpret and present the data.

HiSeq X-Ten genome sequencers (Image from 

Who is buying these advanced machines? When Craig Venter founded his third company, Human Longevity, Inc., he raised $70 million in private funding. It purchased two Illumina ten-machine systems, with plans to buy more. In an effort to extend human lifespan, his company will focus on diseases related to aging, such as cancer, diabetes, obesity, heart and liver diseases, and dementia. It plans to sequence 40,000 complete human genomes yearly. The Broad Institute in Boston, the Garvan Institute in Australia, and Macrogen in Seoul are among other first customers.

"What I'd like to do," Dr. Venter said in a Medscape interview, "is have your genome done as a starting point of medicine — equivalent to having your total blood count and urinalysis done when you go to a hospital."  And he makes it known that hundreds of thousands of individual genomes need to be sequenced in order to have a meaningful database for any analysis or diagnosis. "One genome is basically worthless, but don't tell my mother that," Dr. Venter said. "Without something to compare it with, you can't interpret it. A hundred doesn't get you much better. We need 500,000 to a million genomes." 

Venter has also partnered with the University of California at San Diego to sequence the genomes and tumors of every single patient at the university’s Moore Cancer Center. The goal: to develop and administer targeted drug therapies for each individual based on his or her personal genome.

Pharmaceutical giants like Pfizer and Genetech are getting in on the ground floor. They’ve both partnered with 23andMe to use their 600,000-strong database. Pfizer will analyze 1,000 diseases, starting with Crohn’s, colitis and lupus; Genentech will study 23andMe’s 12,000 Parkinson’s patients along with 1,300 of their parents and siblings. The NIH gave 23andMe $500,000 to study asthma.

But America’s advancements pale in comparison to China’s efforts. According to Al Gore’s book The Future: Six Drivers of Global Change, China has spent $100 billion on genetic research in the last decade, and has more gene sequencers than anywhere in the world. They have more at one institute, the Beijing Genomics Institute, than we have in the entire United States. Gore says the Chinese plan to sequence the genome of every individual in an attempt to understand intelligence and assist in identifying professional aptitudes. Sounds like something Mao would have used in his Cultural Revolution…

Enquiring Minds Want to Know

So, is there value in using today’s consumer-gene-sequencing services to see parts of your genetic code?  For me, the answer is yes, and that’s even before I understood all the caveats. I am an early-adopter. I want to participate in citizen-science, to be part of crowd-sourced genomic analysis. I am happy to give my data to 23andMe’s database for use in correlating SNP mutations with disease.

Are there risks? Some friends say I am nuts to allow the company to have my DNA information, that my greatest concern should be protecting the privacy of my genome. But that horse has left the barn. Google, Facebook, the NSA, even Target, already know everything about me, what difference does it make if my genetic data is available?

Some fear that insurance companies will refuse coverage based on genetics. However, because of the efforts of Francis Collins, the Human Genome Project director, the Affordable Care Act does not permit health insurance discrimination based on genes. California goes even further: genetic discrimination for life, disability and long-term care insurance is forbidden. I’m not concerned about genome-based job discrimination – if applicants were rejected on this basis, employment lawyers would have a field day prosecuting companies that turned down applicants based on a handful of SNP variants.

Online gene sequencing companies could use cookies to correlate my genome with my search history, email, or Facebook habits. 23andMe could plaster my Health Reports with ads. Even worse, companies could sell their data to drug companies, they could aggregate everyone in a certain zip code who is blonde, a good sprinter, needs more antacids than normal, and looks at pictures of baby hedgehogs and kittens. But it still doesn’t bother me, other than the annoyance of creepier pop-up ads.

My personalized ad

Other friends say they do not want to know their risk for diseases with bad outcomes, like Parkinson’s or Alzheimer’s. They may have seen a parent suffer. They may be frightened, or philosophical, preferring to let their future unfold without their knowledge of it. Others don’t want to deal with the uncertainty of a statistical risk. 23andMe does make you opt in to see your results for the particularly nasty diseases. But research examining how people are affected by learning their genetic information shows that they do not become depressed or anxious upon learning their results.

I’m firmly in the camp of wanting to know what risks I may have, even if they turn out to be false-positives. I want to be prepared and take action now if I can. Furthermore, I’m an optimist: there may be treatments for these diseases just around the corner.

My APOE4 Alzheimer’s prediction is interesting to me, since no relatives had dementia. However, my mother died at 60, and her father at 30, so I really can’t know if they would have developed Alzheimer’s when they got older. Perhaps I need to ask Samuel Borg Olssen’s mother if any of her uncles or cousins in Sweden had late-life dementia.

So what do I suggest? In general, I’d say if you want to identify mutation-caused risks, you should use several consumer gene-sequencing services and look at the combined results. Upload your raw data into the various genetic analysis engines. If you are taking drugs you learn you are sensitive to, tell your doctor. Also mention any super-high-risk predictions, particularly if close relatives have the disease. Your doctor can order specific gene-mutation tests that have better disease predictability.

Whats it All About?

So what are the most important things I learned from my adventures in recreational genomics? 

First, the reassuring thought that I am NOT my genes; that I am not simply a piece of hardware with an owner’s manual and known outcomes.

Second, I learned more about how I am connected to my ancestors and relatives, that what we actually share biologically is very little, despite being members of the same tribe. For me, this is a somewhat radical understanding of the meaning of family, and makes me feel a bit lonely.

But then, I learned so much about the biological complexity of life, which becomes more awesome and mysterious to me, with each new, deeper level of scientific discovery.

And finally, I saw how I am more intertwined with my surroundings than I realized. I am part of Odo’s Great Link after all! (I confess to some spiritual awe.)

But I already knew the real and practical outcomes before I embarked on the journey. It has little to do with SNPs, transcription factors, or A’s, G’s, C’s and T’s. It’s simple—it’s what our doctors, our mothers, and philosophers have already told us. Enjoy your friends. Play with your family. Eat healthily – vegetables, fruits, proteins, whole grains, and little processed foods. Exercise and meditate. Be productive in your work and hobbies. Take vacations and have adventures. Laugh. Live in the moment. Choose joy and choose love.

And Carpe Diem.

(Want to start the journey from the beginning? Head on over to Part 1)