Sequencing Your Own Genome! How Can It Be Done? Has Anyone Here Done It?

Question

[JacobS]

Working as a bioinformatician, I am always interested at the increasingly economically-feasible possibility of sequencing my own genome both for fun and medical insight. I wouldn't practice this sort of armchair genetic counselling on others, but knowing the caveats involved I would happily pick apart my own alleles and see what's there. For me, this seems like a logical use of our skills rather than an unusual practice; if I was a mechanic I would certainly work on my own car.

So, has anyone actually done this on themselves? I'm not talking about some minuscule 23andme approach, I would want full genome sequencing with adequate depth to have statistically significant allele identification. I have heard the oft-quoted $5K figure in terms of a sequencing price for a human genome, but what if we can prepare our own DNA extraction and library preps? Which sequencing centers could we use? Would university genomics cores be allowed to perform the sequencing even if we are willing to pay the standard fees? Could we use industry connections to bring the price down a bit? Are there legal implications, such as reporting to insurance companies?

I haven't seen much talk among bioinformaticians on the topic and would love to hear some different perspectives.

[14134125465346445]

BGI has agreed to do it for one family: http://manuelcorpas.com/2012/06/21/crowdfunding-genome-project-day-2-bgi-officially-agrees-sequencing/

[zx8754]

Not worried about conflict of interest?

[JacobS]

Do you mean if I sequence at an academic institution? Or between me and insurance companies?

[zx8754]

with academic institution.

[JacobS]

I certainly would be. That is a large reason I asked this question, to determine how other bioinformaticians go about doing this without utilizing the resources at their fingertips

[Biomonika (Noolean)]

I am really curious about the answers as I already spent some time thinking about this:) I think that the main problem is that DNA extraction and library preparation is not something you can do at home and that even if you are ready to pay all the expenses and send it to sequencing company on your own costs, university administration is not prepared for this. They do not have way how to take your money for this even if you would love to pay them (I would be happy to find out I am wrong about this). And according to insurance companies - why would you report _anything_ to them? With all the possible errors and uncertainties when analyzing such data - could you really be convinced that you are prone to some serious disease? Could insurance company be? Also, I consider DNA to be of purely personal information about you.

[Manuel Corpas]

The way I did the sequencing was to send saliva samples to the BGI using Oragene kits. It took a little bit of convincing from the BGI because at the time they would only do the sequencing from blood samples, which meant it had to be sent with ice and in very special conditions, and how was I going to extract my blood? Unless I did it in Spain with a friend clinician, extracting blood in the UK with no contacts for a purpose like this was completely unthinkable.

I am very grateful to Michelle Mao, the UK sales rep from BGI at the time, who did the negotiations with the BGI and allowed me to do the sequencing using Oragene kits. The BGI did the DNA extraction and the libraries for free.

At that time Oragene could send you 4 free trial test tubes (you should check their website, perhaps they still offer this). Unfortunately by the time I found the money to sequence my family, the DNA stabilising suspension for the Oragene kit had expired, so I asked some friends at the Sanger Institute if they could spare some spit tubes (they were running an internal trial with staff from the Genome Campus to sequence some SNP markers).

Luckily Jeff Barrett and Lizzy Langley at the Sanger let me have a few kits. She kindly posted them to my home door! With Oragene it was really easy, we just spit in the test tubes and sent them to the sequencing company, in our case Hong Kong, which is where everything is sent for the BGI. I was able to send my samples with DHL and I think it cost me ~£80 from the UK.

The picture above shows my hand and the Oragene spit test tubes with my mum, dad and sister's saliva ready to be sent.

The good thing about Oragene kits is that, unlike blood, they do not need to be kept in ice; as soon as you close the tube, a suspension is mixed that stabilises the DNA. So it is clean and easy to transport. Sending my poo sample however, was a totally different story. I can talk about that story in another post if you will ;-)

[JacobS]

Thanks for the thoughts. I completely agree about not reporting to the insurance company! However, there are some legal implications that will need to be addressed directly in the future. For instance, if you smoke, you are legally required to report this to insurance companies, otherwise you may forfeit coverage due to failing to report your accurate medical history. I would keep my sequence completely private, but I wonder how long that will remain a possibility.

[Devon Ryan]

Assuming you're in the US, aren't the health insurance ramifications covered by the Genetic Information Nondiscrimination Act?

[JacobS]

Good point! I certainly hope there won't be loopholes to take advantage of. I think back to the Curry case in 2005-2006: http://sports.espn.go.com/nba/news/story?id=2180298

[Devon Ryan]

Yeah, the devil is going to be very much in the details (or ambiguously written sub-sub-sub-clause)! I should note that GINA is only supposed to be for health insurance. Life insurance and such could still reject you (assuming you told them anything...).

Answer 1

[Manuel Corpas]

Hi there

I have sequenced the exomes of 4 of members of my family, myself included, done the SNP chip with 23andMe and sequenced my personal shitome (or gut microbiota if you will). Because I was doing this with private money and efforts, initially myself and my whole family afterwards, the process has gone through many stages. My exome sequencing I did it in 2011. It cost me $999 which I raised privately with my family support. The raw fastq data are available through figShare:

http://figshare.com/articles/Son_exome_files/92584

The sequencing was done with Illumina HiSeq. As soon as I could, I made these data available through my blog for people to download. I made sure it had a public domain license. The only thing I asked people was that, if they so wished, they feed me back all the interesting findings they encounter. Here is the blog entry:

http://manuelcorpas.com/2012/01/23/my-personal-exome-now-publicly-released/

The response was so overwhelming, that I even got companies like Oxford Gene Technology (OGT) who analysed my exome for free, as they were developing their own internal pipeline. They gave me back the SNPs derived from my exome using GATK and a report which I published in my blog. Looking at the summary metrics in OGT’s report, my personal exome produced:

• 30,702 variations to the reference genome (GRCh37)
• 5,565 non synonymous coding variations with consequences
• A minimum of 61.42% of the on-target regions, covered with a depth of at least 20x
• A total of 2.54 Gigabases of sequence data read and aligned at high quality.

These data are available here (again, in the public domain):

http://figshare.com/articles/SNPs_from_the_Son's_exome_data/92819

I felt that this opportunity to crowdsource the analysis by publishing the data was really effective. I talked with my family through that and we decided as a family to publish and release all our data (with their informed consent). I reported how this process went in a publication in the Journal of Genetic Counseling, "A family experience of personal genomics":

http://link.springer.com/article/10.1007%2Fs10897-011-9473-7

After that, and given that for my family we still did not have sequencing data yet, we started a crowdfunding campaign. The object of this campaign was to raise money to sequence the whole genomes of 5 family members. This cost at the time $20,000 (early 2012). I was able to raise >$3,000, which by some "experts" seemed a failure. This story was picked up by the Science journal:

http://news.sciencemag.org/2012/08/keeping-it-family

With the money raised we were able to sequence 3 exomes of 3 members of my family, my mum, dad and sister. These data were made public as soon as they were received and uploaded them for people to use, in the hope that companies, users and anyone around would help us find interesting things. Again, this was picked up by several companies, interested to test their pipelines or simply wanting to raise awareness of their products. Notably, InSilicoDB did the BWA alignment and GATK variant calling, also including my exome. These data are readily available for analysis through InSilicoDB:

https://insilicodb.org/app/browse?q=ISDB11122

In there you will also find the metagenomics data from my own poo, just in case you are also interested to know. After my initial call for crowdsourcing the analysis of my personal fecal DNA, I had some results sent back via Twitter from Willy Valdivia at Orion Biosciences. Willy kindly sent me a couple of figures with a histogram of the percentage of DNA for top 25 organisms for my fecal sample:

http://manuelcorpas.com/2013/06/10/a-glimpse-of-what-my-fecal-dna-contains/

The VCF file with the variants for the 4 exomes is available here:

http://figshare.com/articles/VCF_file_from_Son/803101

Some people have talked about this sources of data as "The Corpasome" and recently I published an article that described our crowdsourcing efforts of the data available at the time. By then it had become clear how useful this approach could be for understanding the genomic information contained in our personal genomics tests.

http://www.scfbm.org/content/8/1/13

I have produced a video where I explain where this adventure has taken us so far:

http://www.youtube.com/watch?feature=player_detailpage&v=xANVOo0oR04

Finally, I would like to thank Albert Vilella for bringing this BioStar thread to my attention.

[Ben]

Great story! Thanks for sharing

[JacobS]

Very informative! I appreciate the detailed post and ample links. Really gives us some good perspective on the topic.

[Istvan Albert]

Awesome post, great summary and perspective as others mentioned as well.

Answer 2

[Biojl]

What kind of questions specifically would you be interested to answer by looking at your genome?

Right now I wouldn't be very interested, since the 'medical' information you will obtain will probably confuse you rather than give a useful advice you could follow. Sometimes (not always) ignorance is happiness.

[JacobS]

I see what you mean, but I doubt checking specific alleles would really be that confusing for well documented loci. Something akin to 23andme's approach, but with the ability to check any loci, instead of their chosen few. Regarding ignoring my genetic code, I think being ignorant of one's genetic predisposition when the tools are available is akin to refusing medicine when sick! This may not be as true today as personal genomics is in its infancy, but imagine the same statement being made 20 years from now when personal genomics is routine with a physical. I'm sure genetic sequencing will eventually supersede cost-effectiveness compared to time consuming wet lab tests in certain applications - especially when considering pathogen identification.

[Alex Paciorkowski]

Whatever you do, talk with a genetic counselor first. What will you or your family do if you find a mutation in a cancer susceptibility gene? What if you find you have a mutation in NOTCH3 which causes CADASIL? Or what about a variant in a susceptibility gene for Parkinson disease? The list goes on, and requires some thought about the uses of presymptomatic genetic testing. Your comment "I doubt checking specific alleles would really be that confusing for well documented loci" sounds very naive.

"I think being ignorant of one's genetic predisposition when the tools are available is akin to refusing medicine when sick!" is also inaccurate. It is not being ignorant, since there are no medicines to treat or cure the majority of human genetic disorders.

[JacobS]

I see what you mean, but respectfully disagree. To me, that logic is analogous to someone who refuses to go to the doctor when they have signs of Parkinson's or Alzheimer's because "there are no medicines to treat or cure" the disease. There is much to be said about being aware or your situation for management and future planning.

[Istvan Albert]

There is also the issue of quality of life, for me at least. The truth is that the "preventive care" for most diseases (genetic or not) is really simple, live healthy, eat healthy, exercise as much as possible, avoid stress and just be happy.

As the results of the 100K genome project demonstrate every healthy individual harbors around 150 loss of function mutations, many of which are homozygous (see this great writeup for details: http://massgenomics.org/2012/11/human-genetic-variation-1000-genomes.html)

One must ask themselves is there value in pointing out all of these 150 mutations when most of them won't even manifest themselves but may greatly reduce one's quality of life. Today we know too little about of what the genomes of healthy individuals should look like. The argument about medical interventions also may not quite hold - throughout history there were countless instances where medical procedures ended up being more harmful than beneficial - the reason for each of these instances were eerily similar: lack of understanding of what we observe.

But of course I don't want to discourage anyone from looking at their genome - all I am saying that there are valid reason for not wanting to know as well.

[Biojl]

I agree, it will be vital in the upcoming years for medical purposes. And I will probably sequence my own when the right time comes. I simply think we are not still there. And yes, It's cool to have all your variation characterized but I fail to see what kind of answers will you get other than ancestry, which you don't need a full genome, obviously.

Answer 3

[Chris Miller]

You can get quotes for custome sequencing through the Illumina sequencing network: http://www.illumina.com/services/genome_network.ilmn I know Illumina has done fee-for-service type sequencing before.

You might also look into one of the small sequencing companies that are popping up all over the place these days.