Linda Avey is co-founder and CEO of Curious Inc., a company that allows people to analyze and interpret their personal data and share it with their community. Along with Anne Wojcicki, Linda co-founded personal genomics company 23andMe in 2006 and left in 2009 to establish a foundation for the causes and treatment of Alzheimer’s disease.
I asked Linda a few questions on genomics, data and digital health. You can find her on Twitter @lindaavey
With prices increasingly lowering, personal genomics is becoming mainstream. What kind of impact will it have on society when the cost of full genome sequencing is negligible?
Genomic information in a vacuum isn’t all that useful yet, regardless of how low the cost. What’s needed is the combination of genomic profiles with patient data, which will enable discovery of common AND rare variants associated with diseases/conditions/drug response. Once we’ve mapped more of the connections, we’ll be able to tell people more about how their genes may impact their health.
The latest shiny technology to read the genome–or the proteome, or the microbiome–tends to get a lot of attention. Scientists rush to acquire the latest sequencing platform and the race is on to push samples through.This really is the easiest component of genetics research (buy a machine, throw in the DNA and, voila, there’s your sequence).
It reminds me of the ‘Do Re Mi’ song in The Sound of Music: ‘A’s, ‘G’s, ‘C’s and ‘T’s are a lot like Do Re Mi. Strings of letters don’t mean much until we put in the ‘words’. This is the hard part: gathering the messy, complex patient information that may yield important correlations when matched back to the genome.
We’re already seeing books focused on targeted diets for people at higher risk for Alzheimer’s disease. Very exciting possibilities here in the use of personal data.
[/pullquote]Too often, research projects rely on a basic disease diagnosis as the extent of the patient information, and these broad labels aren’t sufficient to tease apart all the underlying complexity we know is there, including varying response to treatments and rates of progression. Bottom line, we still have lots of research to do before personal genomics has broad utility; until this happens, it’s unlikely to go truly mainstream. But, luckily, there’s proof–via 23andMe–that people are willing to join and support crowd-sourced research.
I’ve spoke with a lot of people in the fitness industry lately – personal trainers mainly – and they’re curious to understand how personal genomics can help from a fitness point of view. Can our genomic data help us determine our fitness goals?
Honestly, I don’t think personal genomics will be that important any time soon for the bulk of the population attempting to stay fit. Where it could play a role is in highly competitive sports where even the smallest edge makes a difference. Longer term, personal genomics may be able to direct people to the most efficient form of exercise (focusing on aerobic vs anaerobic work-outs, for example) and strategies to avoid stress on parts of the body that might be prone to injury. But a lot more research has to be done to identify these biomarkers. Until then, we’ll probably see lots of gimmicks, not unlike what we see in the cosmetics industry.
An area where personal genomics could eventually have great utility is in determining individualized diet regimens. We already know there’s no one-size-fits-all when it comes to the appropriate mix of proteins/carbs/fats/nutrients. Factoring in inherited risks of allergies and things like lactose and gluten intolerance can help guide people to the right combinations that optimize their well-being. We’re already seeing books focused on targeted diets for people at higher risk for Alzheimer’s disease. Very exciting possibilities here in the use of personal data.
What’s next in personal genomics? I saw you once said that advertisers could use personal genomic data to market products to us. Can you elaborate on this?
Definitely will be a challenge to make sense of the massive amount of data out there, especially when you consider social media, search, spending, tracking.
[/pullquote]Here’s a scenario: a woman goes to Google to find recipes targeted to her genetic profile that she opts to make accessible for that search. The search algorithm factors in the appropriate markers in her genome known to positively associate with certain ingredients. Retail grocers can then take those ingredients, convert them into recipes and shopping lists and offer discounts for same day delivery (delivered by a self-driving car?). Snapchat-style functionality could then delete any genetic data used so it’s only a temporary share.
What are your thoughts on uBiome? Is it the 23andMe for the human microbiome? What can we expect to learn when they begin analyzing people’s (including mine) microbiomes?
I love uBiome! (Disclosure: I’m also an advisor.) The parallels to 23andMe are numerous. They’re using a similar model to crowd-source the discovery of associations between microbiome profiles and varying health states. It’s the same idea: phase 1 is construction of the research/knowledge database and phase 2 becomes the reference for comparison as a means to assess current health states. The provocative and exciting aspect of microbiome profiling is the apparent ability to manipulate microbial make-up through procedures like fecal transplants.
Your company, Curious Inc., helps people interpret and understand their own health data. What was the rationale behind launching it?
Curious is a natural extension of ideas I had at 23andMe. But now we’re giving total control of the question-asking to the user. With the mad-cap emergence of sensors, devices and other sources of personal data, my co-founders and I see a need to rope it all together in a single place to enable exploration and discovery. It’s a platform/workspace play, which has its challenges but we think it’s the only way people can start to make sense of the mind-boggling amount of data at our disposal.
We are creating more data about ourselves than ever before via activity trackers, smartphones, payment cards etc which is great but will there be any issues in determining how to make meaning from it? Is data interpretation and consultancy an upcoming industry?
Definitely will be a challenge to make sense of the massive amount of data out there, especially when you consider social media, search, spending, tracking. It’s so early in the analytics phase of big personal data, especially in the application of our learnings toward better health. We tend to get ahead of ourselves—we first have to build the infrastructure before we start finding answers. It’ll also be about knowing what to throw out.
Thinking from a commercial angle. Apart from helping us improve our health can we use this data for other means? Can you envisage a world where consumers exchange their personal data commercially to reap benefits? Say, for example, exchanging data for rewards, discounts etc?
I’d like to see an expansion from the Google model (where a company reaps huge rewards from our web behavior) to a more democratic, distributed model. Seems to be a logical evolution—we just need to figure out the technology to scale it.
There is a lot of innovation taking place in health at the moment. Where do you see all of this going? What will the next five years look like?
Five years is probably too short a timeframe to see major change. Things are unfortunately very slow to advance in the health space. I think we’ll continue to see the emergence of disparate pieces of the solution. Maybe ten years from now it will start to coalesce into a more cohesive system that supplants the monolithic, broken model we’re currently living with.