A Very Personal Endeavor
My interest in personal genomics started shortly after I came to work to the Wellcome Trust Sanger Institute, Cambridge, UK, where I was employed for a database that collects genomic and phenotypic data from patients with severe genomic disorders. Work in this database involved most of the time making accessible to clinicians its incorporated genomic tools for analysis. This work made me aware of the need to make accessible the nascent field of personal genomics, not just to clinicians but to the general public as well. If clinicians found it really challenging to interpret genomic information from their patients, patient themselves should find it even harder. I realized how few options are available to the ordinary folk who just wants to understand his/her genes: he/she has to invest a sizeable amount of money to analyze his/her genome and more so if face-to-face/genetic counseling support is needed. At that time there was hardly any literature describing how to communicate and share genomics results with family members.
I was able to perform a personal test using a Direct-to-consumer (DTC) genomic testing provider. When interpreting my own results, I became progressively frustrated by the lack of information that personal genomics reports allowed their customers. I particularly missed being able to navigate and browse through all the raw data upon which my report built its genetic risk predictions. I was predicted a relatively high risk of suffering from prostate cancer (30% chance), yet I had no knowledge of any member of my family being affected by the disease in my paternal or maternal line. This led me to research more closely my 0.5 million personal genomic markers from my DTC test: I wanted to see where my prostate cancer risk factors were located in my genome and the scientific basis of these predictions.
Tinkering with this data I then decided to make my personal DTC data available on the Internet through a web service (DAS), which enabled me to view my data in Ensembl, a popular genome browser. Understandably, I had to face criticism as some coworkers were concerned that my willingness to make my personal data available via the web might compromise their project. I was finally agreed that I would carry out my investigations with my personal genome data on my own, reporting all of my findings independently, which I did in this blog. Since then, I have always tried to present the plain unadulterated truth about what I was doing. This, I think, has been one of the hallmarks of this blog, which with time has become independently recognized as one of the most influential in the field.
Despite being able to view my personal data in Ensembl, I realized that ordinary citizens would not be able to visualize their personal genome data without a great knowledge of bioinformatics and computer science. I realized the need for a personal genomics browser exclusively tailored to non-computer savvy users that would allow visualization of personal genomics data in the context of the chromosomes they are located in, the affected genes and other potential overlapping variations. This would not save the need for users to know about genetics or having to be minimally computer literate but at least it was a step forward on the right direction. With the help of a friend I built arguably the first personal genomics oriented genome browser, myKaryoView. We released it for free and to date it is still open source.
I have openly blogged about my personal genome investigations, making all my data freely available for anyone to download and research on. The following year after I carried out my DTC test, my family followed suit, taking their own personal genomic test and making all the raw data available on the Internet, with no restriction of use. This was another action that raised concern in some circles as some people doubted that my family, not being experts, would be able to have a perfect consent for what they were doing. I have always argued to this that no one’s consent is ever perfect in personal genomics tests because one never knows what s/he is going to find, acceptance of personal genomic risks is more a question of attitude rather than being informed.
All our experiences in analyzing our family genomes have appeared in a series of blog entries and videos and several landmark papers in the field of personal genomics. One paper described my personal experiences when communicating personal genomics results to other members of my family. Here I showed that there were important findings in terms of general attitudes of lay people towards acceptance of their personal genomic risks, the danger of reaching potential eugenic conclusions with risk scores and the current generalized lack of support for a whole family personal genomics. Another recent paper, which I led, described the results of analyzing my genome and the genome of my family with publicly available tools. We were able to report that my sister’s genome is more similar to my father than my mother as she has inherited a deletion from my father that I have not, making her more significantly similar to him.
I also started a Crowdfunding campaign to sequence the whole genomes of my family in which we were able to gather $3,526 USD in 40 days. The whole purpose of this project was to be able to analyze in much more depth the genomes of the family and make both the data and results analyses available through the Internet for public use (with my family’s informed consent). Part of the reason for making this data available is because there are hardly any family data models with which to train and build those tools.
I has never feared risking my reputation by following an unconventional/disruptive way of communicating science. I believe I have been able to stir some people’s opinions as evidenced by the tremendously positive response I have had through Twitter to some of my publications.
I will continue to work tirelessly on the democratization of personal genomics data. Sadly, this is still a dream and not a reality, starting from the fact that there is a catastrophic lack of publicly available tools for interpretation of results. I believe that anyone, regardless of country, race or economic situation should have access to his/her personal genomic data and the ability to research them as much as s/he wants with resources that are public and easy to use. I will consider successful my contribution to this field if this work has raised at least enough to make a difference in this problem.