informatics

Why you CAN have your $1000 genome - so long as you learn what to do with it [Genetic Future]

sb admin — Fri, 07 Jan 2011 09:10:00 +0000

Why you CAN have your $1000 genome - so long as you learn what to do with it [Genetic Future]

As part of his Gene Week celebration over at Forbes, Matthew Herper has a provocative post titled "Why you can't have your $1000 genome". In this post I'll explain why, while Herper's pessimism is absolutely justified for genomes produced in a medical setting, I'm confident that I'll be obtaining my own near-$1000 genome in the not-too-distant future.

Matt's underlying argument is that while sequencing costs will continue to drop, obtaining a complete genome sequence that is sufficiently accurate for medical interpretation will require additional expenses (increased sequence coverage to ensure accuracy, all of the computation required to stitch the raw data into a useable form, and paying doctors to perform the interpretation) that will keep the cost of medical sequencing well above the arbimagical US$1,000 threshold. Instead, Herper argues, we will likely see medical-grade genomes stay above $10,000, or at least above the $2,000 currently forked out for MRI scans.

There's certainly some depressing truth here. I believe Herper is right that if you intend to access your genome sequence via a traditional medical route, it will certainly cost more than $1000 for the foreseeable future - if indeed you can get access to it at all, which is by no means guaranteed. The costs of clinical sequencing will include even more overheads than Herper notes in his short post: for instance, even as the accuracy of high-throughput sequencing technology improves, there will still be a need for variants with major medical impact to be independently validated in clinical labs, and custom assays don't come cheap.

However, many of these extra costs of clinical sequencing will be further inflated by regulatory demands (at least some of which will be arbitrary and pointless), and many will only apply if you obtain your genome through the medical system. Individuals with the motivation to seek alternative routes will be able to obtain a perfectly serviceable genome sequence at a substantially lower price: they'll have to be cautious in how they interpret the results, of course (although, importantly, this is also true of a medical test), but it will be possible to obtain substantial and potentially extremely useful information from your own genome without having to pass through the clinical toll-booths.

The key obstacle will be the development of cheap (or free), intuitive tools for annotating large-scale genetic information. Purchasing the sequencing itself will be trivial: even if the FDA succeeds in crushing innovation in direct-to-consumer genetic testing in the US, there will be plenty of companies abroad (especially in East Asia) willing to convert a mailed saliva sample into an assembled genome sequence. What the individual needs to do with that sequence is to (1) validate that the sequence they receive is in fact their own genome; (2) extract medically useful variants; (3) confirm that these variants are real; and (4) figure out how that information should be used to make health and lifestyle decisions.

Only one of these steps (the final one) will require direct consultation with the medical profession. The first will require comparing the sequence with independent genetic data (such as a genome scan from a company like 23andMe) to check that the two match the same individual (i.e. you), and also to provide an indicator of the global quality of the sequence. The second step is currently extremely challenging, but we can expect tremendous innovation in genome interpretation software and databases of functional variants over the next few years that will gradually simplify and improve this process. The third step will require sending another DNA sample to a company that does affordable, custom assays of a small number of genetic regions of interest using an independent technology. And the final step will involve discussing the results with everyone who might be able to tell you something useful about them, including your family and your doctor.

None of this is simple, but it will become easier with time. As the retail costs of sequencing drops, a substantial niche will develop for innovators providing affordable, intuitive, accurate interpretation tools (embryonic versions already exist: see, for instance, Promethease or Enlis Genomics). Open-source academic software built for large-scale sequencing projects will be adapted for use by non-specialists. The increasing availability of large-scale computing power (for instance, via Amazon EC2), coupled with this intuitive software, will make even compute-intensive analyses available to the educated, motivated lay-person. (Incidentally, tracking and fostering the development of these tools is one of the motivations behind the Genomes Unzipped project - and you'll hear more about our plans in this area in 2011).

Done carefully, there's no reason why a DIY genome couldn't be every bit as useful (or indeed as useless, in many cases) as one obtained through the doctor-as-gatekeeper route. As Dan Vorhaus argued after last year's sample mix-up at 23andMe, it is likely that clever DIY genomicists will be better at picking up certain kinds of errors (such as sample swaps) than clinical labs would. In terms of accuracy, while retail genomes may not reach the same quality standards as those generated by clinical labs, increased competition and innovation in the direct-to-consumer space will mean they're unlikely to lag too far behind - and judicious use of independent validation of important variants would in most cases raise the reliability to a level is equivalent to, or even higher than, a medical-grade test. Of course, for the very small number of genetic variants per genome that might require urgent, serious action - such as BRCA1 breast cancer-associated mutations - individuals can always fork out for a clinical test.

What proportion of people will take this DIY route? This isn't for everyone. Those wealthy enough to blithely fork out $10,000 for a medical genome interpretation, or sufficiently unwell to be able to convince their insurance company or public health system to pay for it, will by and large simply take the expensive medical route. Of the remainder, relatively few people will be sufficiently motivated to develop the background knowledge required to make sense of their genome, even if the analysis software is relatively intuitive. But for the non-trivial fraction of the population who want to know about their genomes, but don't want to pay the inflated costs associated with medical-grade sequencing - and I know this is a category that many readers of Genetic Future fall into - this will be an attractive and feasible option.

In addition, there will be advantages to the DIY approach beyond the lower cost. People who actively engage in the process of constructing useful information from raw sequence data - regardless of how intuitive the software is for doing it - will automatically learn important lessons about the nature of genetics (just as anyone who has given more than a casual glance at their own 23andMe profile has automatically learnt something important about the probabilistic nature of genetic risk factors for common diseases). They will also have opportunities to ask and answer fascinating questions that would be irrelevant in a purely medical consultation about your genome: for instance, what does your genetic information tell you about your ancestry?

They'll be able to ask these questions because they will own the data. How easy do you think it will be to obtain your raw genome sequence from your doctor to use to satisfy your own curiosity? How many forms and disclaimers will you need to sign? How many times will you need to listen to someone tell you that the data files are just too large, that the formats are inaccessible to lay-people, that your request is extremely unusual and will need to be considered for months by a hospital committee in the name of "health data privacy"? Anyone who has ever tried to get access to their own medical records will know how tedious and shrouded in unnecessary mystery this process can be; imagine how much larger the obstacles will loom when the system has the additional excuse of large, complex file formats to throw in your path.

Importantly, your genome is just one contributor to your present and future health. DIY genomics will, I hope, be part of a larger ongoing trend towards individuals taking greater personal responsibility for tracking and maintaining their own wellness - a task, incidentally, which modern healthcare systems are spectacularly ill-equipped to perform, something that seems unlikely to change substantively in the near future. As Western populations age, this broader shift of responsibility will be essential for healthcare systems to survive.

But I digress. My point here is simply this: Herper is perfectly correct that the overheads imposed (for a mixture of valid and arbitrary reasons) by medical-grade testing will ensure that clinical genomes remain expensive. But for those of us willing to learn the skills required to go outside the system, the $1000 genome is rapidly approaching. We just need to be ready to make the most of what it contains - and to reap the benefits of accessing that information as an active, engaged participant rather than a passive recipient.

sb admin Fri, 01/07/2011 - 04:10

Categories

Free Thought

Why you CAN have your $1000 genome - so long as you learn what to do with it

dgmacarthur — Fri, 07 Jan 2011 09:10:00 +0000

Why you CAN have your $1000 genome - so long as you learn what to do with it

dgmacarthur Fri, 01/07/2011 - 04:10

Categories

Free Thought

Bioscience Resource Project critique of modern genomics: a missed opportunity [Genetic Future]

sb admin — Wed, 15 Dec 2010 06:15:00 +0000

Bioscience Resource Project critique of modern genomics: a missed opportunity [Genetic Future]

Late last week I stumbled across a press release with an attention-grabbing headline ("The Causes of Common Diseases are Not Genetic Concludes a New Analysis") linking to a lengthy blog post at the Bioscience Resource Project, a website devoted to food and agriculture. The post, written by two plant geneticists, plays a tune that will be familiar to anyone who has encountered the rhetoric of GeneWatch UK: basically, modern genomics is pure hype perpetuated by scientists seeking grant money and corporations seeking to absolve themselves of responsibility for environmental disasters.

The post is long, but its core argument can be summarised as follows:

Genome-wide association studies (GWAS) have failed to find variants explaining much of the risk of common diseases like type 2 diabetes;
The potential hiding places postulated for the remaining "missing heritability" are implausible;
Many epidemiological studies have shown a major role for environmental factors in determining disease risk;
Studies estimating the proportion of disease risk determined by genetics using twin pairs are flawed;
Both corporations and medical researchers have incentives to prop up the notion that common diseases have genetic causes;
Therefore, the notion of major genetic causation for common diseases is a fallacy, and we should stop looking for disease genes in favour of investing in beneficial environmental changes.

These claims would be fascinating, if true. However, while the article makes some (scattered) valid points, its central claim (that the results of GWAS suggest that genetics plays little or no role in the causation of common diseases) is entirely false, and the authors rely on a combination of distortions and statistical misunderstandings to make their case.

Unfortunately the article has not simply lapsed back into the internet obscurity it deserved: over the weekend a link to the article was posted on Twitter by popular author Michael Pollan, bringing it to the attention of his ~40,000 followers. Pollan's tweet and the cheer-leading responses from his followers were subsequently picked up and blasted over at OpenHelix, leading to an exchange with one of the authors in the comments. The article was also criticised for a schoolboy statistical error by Luke Jostins, but received a qualified positive review from Mike the Mad Biologist.

So, let's take a closer look at how well the some of the claims in the article stand up.

Why was the post written?

The article itself is written in a reasonably neutral tone, which could easily fool the casual reader without a solid background in genetics (like, perhaps, Michael Pollan) into seeing it as a dispassionate critique of the field. However, it's important to read the post in the appropriate context.

In a comment over at the Huffington Post found by Keith Grimaldi, one of the authors explains the key messages and motivations of his analysis:

We have just reported that genetics now demonstratÂes that genes cannot be the cause of common diseases:

http://wwwÂ.biosciencÂeresource.Âorg/commenÂtaries/artÂicle.php?iÂd=46

That means environmenÂt must be the entire cause of ill health, i.e. junk food, pollution, lack of exercise, etc. The reason we wrote an article about human genetics (when we are a food and agriculturÂe website) is that we believe that if people live right, agriculturÂe and therefore the planet will more or less fix itself. [my emphasis]

This quote is illuminating in a number of ways. Firstly, it shows that there is no nuance in this argument: the authors aren't attempting to argue that genes play a smaller role in common disease than geneticists expected, but rather that genetics plays no role whatsoever.

Secondly, it reveals the motivations behind the post: the authors have assembled this critique, despite their acknowledged lack of expertise in the field, because they want to encourage a greater focus on behavioural and economic changes to bring large-scale environmental benefits. A noble cause, to be sure, but not one that necessarily encourages them to take a balanced approach to the discussion.

I don't mean to discount the post itself on the basis of its authors' motivations, but I do think it is important to read the piece in this context.

OK - on to some of the specific claims made in the piece.

Possible explanations for the missing heritability are post hoc and implausible

The authors claim:

A problem for all these hypotheses, however, is that anyone wishing to take them seriously needs to consider one important question. How likely is it that a quantity of genetic variation that could only be called enormous (i.e. more than 90-95% of that for 80 human diseases) is all hiding in what until now had been considered genetically unlikely places? In other words, they all require the science of genetics to be turned on its head. [italics in original]

This is complete nonsense. Indeed, the authors' question should be turned on its head: How likely is it that a technology that we know is only well-powered to find risk-associated variants that are common and have reasonable effect sizes will have found all - or even most - of the variants underlying common disease risk? If the answer to that question is "not very likely" - as it clearly is - then the authors' argument falls apart. Genome-wide association studies (GWAS) were not conducted because scientists expected them to find every disease-associated variant, but because they were a place to start with the technology that was available; the fact that a large fraction of the heritable risk remains undiscovered is not a sound reason to doubt that risk was heritable in the first place.

Some fraction of the missing heritability for complex diseases may turn out to lie in exotic candidates such as epigenetic inheritance or heritable variation in microflora, but these aren't yet required explanations. There are also perfectly mundane locations that haven't yet been explored by modern genomics, and would require absolutely zero changes to "the science of genetics" to investigate. For instance, genome-wide association studies (GWAS) conducted to date have been seriously under-powered to detect risk variants at low frequency (less than 5%) in the population, as well as common variants with individually very small effects on disease risk - yet there's no reason not to expect an appreciable fraction of the population variance in disease risk to fall into these categories. Or, again, are we expected to believe that the distribution of allele frequencies and effect sizes for disease risk variants falls entirely within the range for which GWAS conducted to date have been 100% powered to detect them?

We haven't even begun to make the most of risk variants we have already uncovered. GWAS are capable of flagging up a region of the genome linked to a disease, but typically don't immediately identify the precise genetic change responsible for that association. More detailed analyses of risk-associated regions (known as fine-mapping) allow researchers to zoom in on variants that are more tightly linked with the underlying causal change - and this alone can substantially increase the fraction of variance explained.

Variants discovered by GWAS are useless

The authors argue:

For each disease, even if a person was born with every known 'bad' (or 'good') genetic variant, which is statistically highly unlikely, their probability of contracting the disease would still only be minimally altered from the average.

Erm, no. Luke Jostins has a very handy post showing the distribution of risk prediction scores for individuals with different combinations of genetic variants associated with three common diseases: type 1 diabetes, type 2 diabetes, and Crohn's disease. Given he'd gone to all the work of collating these distributions, I asked him to do precisely the analysis the post authors describe here, and compare the predicted risk of individuals with all possible risk variants to the population average.

Here are the results for people with the average risk vs those with the highest number of risk variants:

Type 2 diabetes: 19.6% vs 41.3%

Type 1 diabetes: 1% vs 65%

Crohn's disease: 0.4% vs 99.6%

This analysis includes only variants identified by GWAS, but it's also based on a somewhat out-of-date catalogue of variants - so updating the results would increase this spread slightly further. [Explanation above edited to correct minor error in original version, which stated numbers were for lowest vs highest risk rather than average vs highest risk.]

Do the authors genuinely believe that the difference between 0.4% and 99.6% risk represents "minimal alteration", or have they just not bothered to actually look into these numbers themselves?

Strong environmental effects on disease risk argue against strong genetic effects

This argument pops up in a number of places in the article. For instance, the authors point out the apparent contradiction between twin studies suggesting that the risk of myopia is 80% heritable, whereas individuals moving from non-Western to Western countries can go from a prevalence of myopia of 0% to 80%. How can these two figures be reconciled?

The answer is that heritability is a number that applies to a specific population within a specific environment. Within white Europeans living in Western countries, who face a reasonably uniform set of environmental risk factors, around 80% of the risk of myopia is genetic. That number will obviously not apply to a population in which some individuals are moving from a low-risk to a high-risk environment, in whom the majority of the risk is primarily determined by that massive environmental difference. However, importantly, that doesn't mean the heritability estimate isn't correct for white Europeans: it just means that it shouldn't be extrapolated to other populations subject to different combinations of genetic and environmental risk factors.

There is no contradiction here, just a misunderstanding of the concept of heritability. The authors' misunderstanding should remind us of the caution that needs to be applied when thinking about heritability, and also that the existence of strong genetic predispositions to common diseases doesn't mean that environmental interventions can't be extremely effective. However, it's not a valid critique of the heritability estimates generated for common diseases.

The evidence for disease heritability from twin studies is flawed

The authors claim:

Studies of human twins estimate heritability (h²) by calculating disease incidence in monozygotic (genetically identical) twins versus dizygotic (fraternal) twins (who share 50% of their DNA). If monozygotic twin pairs share disorders more frequently than do dizygotic twins, it is presumed that a genetic factor must be involved. A problem arises, however, when the number resulting from this calculation is considered to be an estimate of the relative contribution of genes and environment over the whole population (and environment) from which the twins were selected. This is because the measurements are done in a series of pairwise comparisons, meaning that only the variation within each twin pair is actually being measured. Consequently, the method implicitly defines as environment only the difference within each twin pair. Since each twin pair normally shares location, parenting styles, food, schooling, etc., much of the environmental variability that exists between individuals in the wider population is de facto excluded from the analysis. In other words, heritability (h²), when calculated this way, fails to adequately incorporate environmental variation and inflates the relative importance of genes. [my emphasis]

As Luke Jostins has already explained at length over at Genomes Unzipped, this criticism is based entirely on a statistical misunderstanding of the methodology behind heritability studies. In fact, the sentence highlighted in bold above is completely wrong: twin-based heritability estimates use between-family variability, not within-family variability, to estimate the proportion of variation that is due to the environment. This misunderstanding completely undermines their argument against heritability estimates.

As Luke notes, there are valid reasons to be cautious about heritability estimates from twin studies - but this isn't one of them.

What this piece could have been

Mike the Mad Biologist has a post about this article, in which he describes it as having "good and bad points". I should also be charitable: although the central argument of the post (that results from GWAS suggest that genetic factors have little or no role in common disease) is completely wrong, there are valid criticisms of the excessive value that is sometimes placed on genetic versus environmental explanations of morbidity.

Stripping away the conspiracy-mongering and accusations of genetic determinism among geneticists (seriously, how can anyone working on complex diseases be a genetic determinist?), there are some nuggets of truth in the article's discussion:

The last fifteen years, coinciding with the rise of medical genetics, have seen unprecedented sums of money directed at medical research. At the same time, research on pollution, nutrition and epidemiology has not benefited in any comparable way.

[...]

This same mindset is accurately reflected in the media where even strong environmental links to disease often receive little attention, while speculative genetic associations can be front page news.

Even as a direct beneficiary of money thrown at medical genetics over the last five years, and someone who blogs entirely about news in the genetic domain, I freely acknowledge that these criticisms have merit. Genetic dissection of common disease is valuable, and will be (and indeed already has been) fruitful in generating new therapies, but it is nonetheless true that research into environmental risk factors and interventions to minimise morbidity is woefully under-funded and under-reported relative to its potential benefit.

This article could thus have been a considered, balanced and valuable critique of the imbalance in funding between research into the genetic and environmental contributors to common disease. Instead, the authors have undermined their argument by wandering into territory they don't understand, and taking an extreme position that is inconsistent with the available evidence. Perhaps they felt that polarising the debate was the only way to get attention - and indeed that approach seems to have worked - but that has come at the cost of destroying the credibility of their message. This was a missed opportunity.

sb admin Wed, 12/15/2010 - 01:15

Genomes Unzipped reader survey [Genetic Future]

sb admin — Fri, 03 Dec 2010 08:45:00 +0000

Genomes Unzipped reader survey [Genetic Future]

A reminder to anyone who reads my other blog Genomes Unzipped that we have a reader survey underway there now, which includes some questions about genetic testing experiences and attitudes towards genetics. We're closing the survey to responses this weekend, so if you're an Unzipped reader but haven't had a chance to fill in the survey, please do so now.

sb admin Fri, 12/03/2010 - 03:45

News from 23andMe: a bigger chip, a new subscription model and another discount drive [Genetic Future]

sb admin — Wed, 24 Nov 2010 07:45:00 +0000

News from 23andMe: a bigger chip, a new subscription model and another discount drive [Genetic Future]

Update 30/11/10: 23andMe has extended their 80% discount until Christmas, without a need for a discount code.

Personal genomics company 23andMe has made some fairly major announcements this week: a brand new chip, a new product strategy (including a monthly subscription fee), and yet another discount push. What do these changes mean for existing and new customers?

The new chip

23andMe's new v3 chip is a substantial improvement over the v2 chip that most current customers were run on (the v2 was introduced back in September 2008). Firstly, the v3 chip includes nearly double the number of markers across the genome, meaning that it is able to "tag" a larger fraction of common genetic variants ("tagging" means that a marker on the chip is sufficiently highly correlated with other markers that it can be used to make a reasonable guess about someone's sequence at those other markers). Secondly, the chip now includes additional custom markers targeting specific variants that the company thinks will be of interest to its customers.

The technical details: the v3 chip is based on Illumina's HumanOmniExpress platform, which includes 733,202 genome-wide markers. The company has also added around 200,000 custom markers to the chip (vs ~30,000 on the v2 chip). We don't yet have full details on what those custom markers are, but there's a summary of the improvements over the v2 chip in the press release:

Increased coverage of drug metabolizing enzymes and transporters (DMET) as well as other genes associated with response to various drugs.
Increased coverage of gene markers associated with Cystic Fibrosis and other Mendelian diseases such as Tay-Sachs.
Denser coverage of the Human Leukocyte Antigen region, which contains genes related to many autoimmune conditions.

Deeper coverage of the HLA is particularly welcome - variants in this region are very strongly associated with many different complex human diseases (including virtually every auto-immune disease), and the v2 chip was missing several crucial markers.

The addition of more rare variants associated with Mendelian diseases like cystic fibrosis is entirely unsurprising, but the devil will be in the details: in the arena of carrier testing 23andMe is up against the extremely thorough and experimentally validated platform offered by pre-conception screening company Counsyl. It will be very interesting to see the degree to which 23andMe focuses on the carrier testing angle in their marketing of the v3.

More power for imputation

From the perspective of those of us simply interested in squeezing as much information as possible out of our genetic data, the v3 chip is a welcome arrival. The additional markers present on the chip will substantially improve the power of genotype imputation - that is, making a "best guess" of our sequence at markers not present on the chip using information from tagging variants.

The HumanOmniExpress platform has some decent power here: in European and East Asian populations, 60-70% of all of the SNPs with a frequency above 5% found in the 1000 Genomes pilot project are tagged by a marker on the chip (in this context, "tagged" means "has a correlation of 80% or greater"). In effect, that means that being analysed at the one million markers on this chip allows you to make a decent inference of your sequence at around another 4.5 million other positions in your genome.

At the recent American Society of Human Genetics meeting, 23andMe presenter David Hinds suggested that the medium-term future for 23andMe rested not in moving to sequencing, but rather on expanding the role of genotype imputation. The new chip will certainly help with that. However, it's worth emphasising that imputation is not a replacement for sequencing: it is only accurate for markers that are reasonably common in the population, meaning that it will miss most of the rare genetic variants present in your genome.

However, improved imputation with the extra markers on the v3 chip will mean that 23andMe should be able to do a decent job of predicting customer genotypes at the positions we currently know the most about - those arising from genome-wide association studies of common, complex diseases. I expect that many customers will see changes to their disease risk profiles as a result of the move to the new chip.

Over at Genomes Unzipped, we've already been looking at various approaches to imputation from our 23andMe v2 data, and we'll put a post together soon looking at how this will improve with content from the v3 chip.

The new product strategy

There are two interesting things that 23andMe has done with the new product line: firstly, it has reversed the transient division of its products into separate Health and Ancestry components; and it has introduced a subscription model in which customers pay $5/month for updates to their account as new research findings become available (previously, customers paid a flat purchase fee and were then entitled to free updates).

The recombining of the Health and Ancestry products into a single Complete package is an extremely interesting move. As Dan Vorhaus notes, the previous separation of the two product lines was plausibly interpreted as a way for the company to pre-empt the possibility of a regulatory crackdown by the FDA: if regulators hammered the company's ability to offer health-relevant tests directly to consumers, 23andMe could easily switch to its Ancestry product to maintain a revenue stream.

In the currently uncertain regulatory environment, the decision to reverse this division is an unexpected one. It certainly appears that 23andMe - flush with cash following a successful $22M funding round - is somewhat more confident than I am about the regulatory future for health-relevant genetic tests; I hope that confidence turns out to be warranted.

Subscription fees: good for customers

The decision to add a subscription fee may prove unpopular with customers (and has already received a qualified thumbs down from blogger Dienekes, albeit for perfectly sensible reasons). However, a business model based on providing continuous product updates that customers don't pay for has never really looked like a viable long-term business model.

I personally see a subscription model as a positive move: it provides a steadier revenue stream for personal genomics companies, which means less focus on splashy discount drives. It also provides more of a financial incentive for the company to improve the ongoing experience of customers: under the current deal customers are locked in for the first 12 months, but after that 23andMe will need to convince them that it's worth continuing to pay for additional content and features.

Other personal genomics companies (e.g. Navigenics) have long relied on some form of a subscription model, but typically at a higher cost. I think 23andMe is hitting a pretty reasonable price point here: I suspect $60/year would be seen by most customers as a fair price.

OMG discount!

That doesn't mean that 23andMe has abandoned the discount drive approach just yet, of course: they're currently offering v3 kits for just $99 (vs the retail price of $499), which must be purchased along with the previously mentioned 12-month subscription fee of $60. Non-US customers can also expect a ~$70 postage fee, based on comments on Twitter.

Anyone who missed out on the DNA Day sale and is keen to take advantage of the v3 content would be well-advised to get in quickly. The discount code is B84YAG.

sb admin Wed, 11/24/2010 - 02:45

Categories

Social Sciences

Why I'm releasing my genetic data online [Genetic Future]

sb admin — Tue, 12 Oct 2010 06:45:00 +0000

Why I'm releasing my genetic data online [Genetic Future]

Back in June I launched a new blog, Genomes Unzipped, together with a group of colleagues and friends with expertise in various areas of genetics. At the time I made a rather cryptic comment about "planning much bigger things for the site over the next few months".

Today I announced what I meant by that: from today, all of the 12 members of Genomes Unzipped - including my wife and I - will be releasing their own results from a variety of genetic tests, online, for anyone to access. Initially those results consist of data from one company (23andMe) for all 12 members; deCODEme for one member; and Counsyl for two of us (my wife and I). As the project proceeds, we plan to obtain and release the results from a far wider range of genetic tests, up to and including complete genome sequences.

In all, the group is currently releasing over 7 million pieces of genetic data mined from our own genomes. Anyone can download the data in raw form, or view it on a custom browser that two of the group assembled using the open-source JBrowse software. Already the data is being used: blogger Dienekes yesterday published an analysis of our ancestry using his own program, EURO-DNA-CALC.

We have plenty more planned over the next few weeks, including discussion of the ethical issues associated with releasing data publicly, especially given the potential impact on family members. We'll also be presenting analyses of our own data: many of us are active researchers in genetics, and relish the opportunity to apply our research tools to our own genomes. We'll be releasing software code allowing others to run the same analyses on their own data.

So, why on Earth are we doing this?

I summarised some of the key motivations for members of the group in my Unzipped announcement post:

we want to share the results of scientific analysis of our own genomes, and as proponents of open data access most of us believe that doing good science means releasing complete data for others to investigate;
we hope that releasing our data publicly will help to guide useful discussions about genetic privacy and the benefits, risks and limitations of genetic information in general;
many of us believe that the ideal resource for genetic research is large open-access, non-anonymous research databases such as the Personal Genome Project, and that sharing linked genetic and trait information openly with the wider community is a public good - and we hope that our own experiences will encourage others to participate in open research projects;
we all believe that many of the fears expressed about the dangers of genetic information are exaggerated, and see this project as an opportunity to have a constructive public discussion about the truth behind these fears;
given the ease with which a dedicated snoop could obtain genetic information surreptitiously (via shed skin, hair or saliva, for instance), some of us argue that the whole notion of genetic privacy is illusory anyway - while releasing our data online makes it easier for people to get hold of it, this is a difference of degree rather than kind.

I wanted to spend a bit of time here expanding on that third point, as this is probably my own primary motivation for engaging in the project.

Any researcher working in genetics or genomics will be all too familiar with the cumbersome bureaucratic obstacles associated with subject privacy and anonymity. Under the traditional research model subject anonymity and data privacy must be protected fiercely, and that leads to substantial hurdles in two key areas: firstly, data sharing between researchers is hindered by the need to ensure that data privacy is maintained; and secondly, layers of protection on subject anonymity mean it is extremely difficult to return research results to participants, even when those results might have health implications.

This is not to say that huge advances in data access have not been made over the last decade, particularly in the field of genomics. Both individual researchers and funding bodies (notably the Wellcome Trust and NIH) have done a commendable job of ensuring that many large genomics data-sets are made available to other researchers through large databases and data access agreements.

However, can we go further? Researchers such as George Church advocate a bold alternative model: recruit research participants who are willing to share their data completely openly with the world. Find large enough numbers of people willing to sacrifice their privacy for public good, and you suddenly have an amazingly powerful resource: a data-set that can be analysed by any researcher in the world with access to the internet, including participants who can play an active role in the research process.

It can't be emphasised enough just how powerful such a resource would be. Right now, virtually all human genetic and medical data is effectively locked away behind tight consent agreements. That means a given data-set only has a certain number of eyes passing over it, with a restricted circle of expertise; one cohort's data might contain valuable insights into the mechanisms by which cholesterol affects heart disease, but if the researchers holding the keys are eye specialists those will probably never be uncovered.

Science moves fastest when people from diverse backgrounds are allowed access to rich data-sets. The closer we hew to the traditional model of tightly restricted access to human data, the slower we will uncover the associations we need to move into the era of personalised, evidence-based healthcare.

Are there enough people in the world willing to forego their privacy in the name of science? That remains to be seen, but flagship studies like the Personal Genome Project - which seeks to recruit 100,000 volunteers willing to share their genomes and clinical data with the world - are already suggesting that this number is far higher than many would have expected. However, visceral opposition to the idea of releasing such information - based often on an exaggerated sense of the power of genetic data, or its potential for abuse - continue to hold sway over the vast majority of the public.

We're under no illusions here: the data from the 12 of us in Genomes Unzipped aren't in and of themselves of tremendous scientific value. However, if we can get people starting to think about the genuine public good that can be achieved by sharing their data with science, and to weigh that good against a realistic sense of the potential harms, then the project has been a success.

Edited 13/10/2010 to clarify that major progress has been made in data-sharing agreements over the last decade, especially in genomics - I apologise to anyone who interpreted my views as minimising the work that has been done in this area.

sb admin Tue, 10/12/2010 - 02:45

Categories

Medicine

The informatics of new sequencing technologies

dgmacarthur — Wed, 16 Sep 2009 05:00:00 +0000

The informatics of new sequencing technologies

Kevin Davies from Bio-IT World has two interviews up today relevant to new DNA sequencing technologies.

Firstly, there's an excellent interview with Clive Brown, vice president of development and informatics for Oxford Nanopore Technologies - one of the most promising contenders in the rapidly evolving third-generation DNA sequencing market. Brown is renowned for his plain-speaking approach (in the article he's described as "the most honest guy in all of next-gen sequencing", a quote originating from David Dooling), and that shows here in his blunt comments on the recent Helicos genome publication (covered here), the technical and marketing challenges faced by new seq tech companies, and the process of poaching employees from the Sanger Institute.

Secondly, Kevin Davies interviews David Dooling, who heads informatics at The Genome Center at Washington University at St Louis (and also writes an excellent blog, PolITiGenomics). Dooling explains why Wash U. chose to invest in the Illumina platform rather than SOLiD, advocates the use of open source software for large-scale genomics, and discusses the transition to third-generation technologies looming on the horizon.

Meanwhile, for those fortunate enough to have subscriptions, GenomeWeb has a dense crop of interesting articles just out including an intriguing investigation into the real costs of sequencing a human genome. The article notes a surprisingly wide range of cost estimates even for providers employing the same technology (Illumina), especially once frequently under-appreciated costs such as detailed downstream informatic analysis are factored in.

Thus, while Illumina is now offering genome sequences at a retail price of $48,000, Stephen Kingsmore, president and CEO of the National Center for Genome Resources, argues that the true cost of a very high-quality finished sequence is closer to $250,000. Kingsmore argues that a large chunk of this cost comes from the current need for intensive manual analysis of the resulting sequence, a cost that will decline precipitously as sequencing technology and informatics solutions improve.

Subscribe to Genetic Future.

Follow Daniel on Twitter.

dgmacarthur Wed, 09/16/2009 - 01:00