Saturday, August 31, 2013

The Most Ridiculous Myth About Cancer Is...

David Robert Grimes, at The Guardian:
"Sharks don't get cancer!" is a factoid that has burnt its way into the public consciousness to such an extent that nothing short of a cultural lobotomy can erase it. The perceived immunity of this predator to cancer has led to its slaughter in order to harvest its allegedly curative cartilage; not only is this no good for sharks, it's no good for humans either.
Sharks do get cancer – indeed, pretty much all complex multicellular organisms do, from dogs to elephants.
I've never heard this one and wish I never had.  Sure, it's possible to pass off the phrase as a hypothesis, but a simple search of PubMed unearths examples of sharks getting melanoma, liver and testicular cancer, as well as reports of many more in the Registry of Tumors in Lower Animals.

Being naive to this rumor, I originally thought it was related to the current media frenzy around "Sharknado", but no, the shark-cartilage/liver-oil-panacea meme has been floating around for over twenty years.

Here's a more comprehensive article at the Journal of the National Cancer Institute about the issues surrounding shark cartilage, its use as an anti-cancer agent in clinical trials, and the companies that sell dietary supplements fabricated from shark tissues.  Even then (in 2005), it didn't seem like shark cartilage had a fighting chance against data it wasn't really useful in a clinical setting.

Friday, August 30, 2013

Taking A PhD From The Bench to The United Nations

Marga Gual Soler shares her career trajectory at Science Careers:
In February, I joined the NGO Branch of the United Nations Department of Economic and Social Affairs for a 6-month internship exploring how science, technology, and innovation can drive sustainable development. ... 
It has been fascinating to discover how science can benefit society at multiple levels: by informing the sustainable development agenda, improving access to education and healthcare, and fostering technological cooperation among countries. I also came to realize that very few people know that access to scientific advancement and its benefits is recognized as a universal human right.
Marga's story is a great example of how almost every 'extra-curricular' activity during graduate school can help develop skills for a future career.  Don't let anyone convince you that the only things that should be done during graduate school are reading, research, and writing; those people don't have your best interests in mind.

If you have a passion to do something unusual or atypical when you're not working - do it!  Taking breaks from your main projects paradoxically increases your productivity.

Thursday, August 29, 2013

A Peek Inside The New Hospital for Sick Children's Research Tower

Tom Blackwell, National Post, reports on the soon-to-be open Peter Gilgan Center for Research and Learning tower, informally known as "the Sick Kids' Research Building":
Scientists have traditionally toiled in small, self-contained labs, mixing mainly with those in the same narrow discipline and doing much of their communication via the Internet, said Janet Rossant, Sick Kids’ chief of research.
That fractured culture is not helped at the Toronto hospital, a major research centre as well as one of the country’s foremost hospitals for children, by having scientists divided among six buildings. All 2,000 research staff will be moving into the new, 21-storey Peter Gilgan tower next month.
It's a nice space (I especially like the atrium), and one which looks like it'll promote teamwork and communication.  The article closes with this bit of sane advice from a scientist:
Ms. Josselyn, a neurophysiologist, said she believes other workplaces can benefit from the human interaction that brought the two scientists together.  “I think it’s always hugely important,” she said.
“If you encounter some problems or you just want to bounce an idea of somebody, it’s so much easier if you already have a connection with them … I think non-work-related interacting with colleagues always promotes work.”
Here are more photos from Diamond Schmitt, the architectural firm behind the tower's design.  If any readers have more to share please send the link to

Wednesday, August 28, 2013

Genomics XPrize Cancelled: Next Gen Sequencing Renders Goal Obsolete

Peter Diamandis, Chairman of the X Prize Foundation, explains how industry completely blew the Archon Genomics XPRIZE goal out of the water:
Back in 2006, we announced the Archon Genomics XPRIZE presented by Express Scripts, which offered $10 million to the first team that could rapidly and accurately sequence 100 whole human genomes to a standard never before achieved at a cost of $10,000 or less per genome. 
What we realized is that genome sequencing technology is plummeting in cost and increasing in speed independent of our competition. Today, companies can do this for less than $5,000 per genome, in a few days or less - and are moving quickly towards the goals we set for the prize.
All things considered, it's completely logical to pull the competition as it's no longer competitive and the prize became a potential hand out.

To get a general idea of the rationale behind the decision, look at these two summary figures at the NIH which show how much the cost of sequencing fell off a cliff in 2007, just after the XPRIZE was announced.  In 2006, the cost per genome was $10 million and the XPRIZE was justified.  Not so in 2013.

So while Diamandis implies that there was a failure in setting the Genomics XPRIZE, I have to disagree.  This was simply a case of an unforeseen technology surprising everyone involved.  That it forced them to adapt in response isn't a win or a fail, it's the normal course of business.

5 Most Overrated Degrees

No, a PhD is not one of them.

Tuesday, August 27, 2013

Life Technologies Gets Into the Brewing Business

Life Technologies Corporation today announced the expansion of its food safety portfolio by partnering with PIKA Weihenstephan to provide molecular testing solutions for the detection of spoilage organisms in the brewing and beverage market. Life Technologies is now the only supplier to provide both enrichment media and PCR kits designed to suit the changing needs and challenges in beverage quality and safety. 
And in doing so, Life helps address a huge problem:
For generations, brewers have been looking for ways to eliminate beer spoilage bacteria, which can impact the smell, taste and clarity of the end product. By partnering with PIKA Weihenstephan, Life Technologies will offer, not only the testing solutions, but also the consultancy services for brewers to understand and respond to positive test results.  
This is a great use of biotechnology!  Here's the full press release at the Wall Street Journal.

Friday, August 23, 2013

10 Tips for Grad Students (Or Anyone) at Scientific Conferences

Here's a nice summary of conference tips by David Shiffman at Southern Fried Science.

By far the best tip (that's coincidentally not the easiest one to execute):
4) Prepare an “elevator speech”. 
If you think it's hard to condense years' worth of work into 30 minute talks, try condensing your thesis topic into 30 seconds!

Thursday, August 22, 2013

GraphStream Dynamic Graph Library

If you're into network graphics and data visualization, check out the GraphStream Project:
GraphStream is a graph handling Java library that focuses on the dynamics aspects of graphs. Its main focus is on the modeling of dynamic interaction networks of various sizes.
And it has several useful features:
GraphStream allows to store any kind of data attribute on the graph elements: numbers, strings, or any object. Moreover, in addition, GraphStream provides a way to handle the graph evolution in time. This means handling the way nodes and edges are added and removed, and the way data attributes may appear, disappear and evolve.
Watch the demo video for example of how graph data can be animated.

Tuesday, August 20, 2013

Protecting Consumers That Buy Direct-to-Consumer Genetic Tests

At Science Translational Medicine, Effy Vayena and Barbara Prainsack discuss the implications of a recent European report on Direct To Consumer (DTC) genetic testing that's overwhelmingly conservative. They point out that:
Potential harm to users appears to be the backbone of the argument for restrictive DTC regulation, little evidence supporting this claim is provided. In fact, recently published research on this issue, although inconclusive, did not report harmful outcomes in the studied cohorts.
The two studies cited point out some evidence that seems contradictory: one study mentioned that users of direct-to-consumer genetic test don't experience distress over test results, and the other that consumers are anxious over results prior to receiving them (understandably so), but after receiving their results, still want to review the results with a health professional (understandably so).

Overall, it seems that your average consumer ordering a DTC genetic test goes from slightly apprehensive before the test results come in to experiencing a need for a helping hand afterwards. 

Whether this is due to consumers receiving results that are mundane in their report (You have an allele carried by 2% of the population that's related to metabolism of sugar), or just being unable to interpret the report in the first place isn't very clear (You have a A>T polymorphism somewhere on Chromosome 19), it seems that consumers aren't as dumb as regulators assume they are:  Once consumers determine that they can't interpret genetic reports on their own they decide to find a pro that can.

But Vayena and Prainsack also make a pretty good distinction of the niche DTC tests are likely to occupy:
The term “beyond-the-clinic (BTC) genetics” better describes what is now known as DTC genetic testing. Besides serving as a more encompassing label under which the various forms of interaction among service providers, users, and medical professionals can be subsumed, the concept of BTC genetics also captures the notion that the boundaries between genetics in and outside of the clinic are blurring (3). It also hints at additional uses genetic information may have outside the clinic that contribute to a broader understanding of well-being that transcends clinical definitions of health. 
Taking this together with the studies trickling in, you might argue that regardless of whether or not consumers end up doing anything with direct-to-consumer test results, little to no harm can be done.  Most people might be motivated to take an greater interest in their own health and even consider taking their test results to a professional for some better guidance.  This, in general, is a Good Thing, right?

Yet there's a peril glossed over at the end:
The report rightly proposes tighter regulation of the advertisement of DTC genetic tests to consumers. The objective of this mandate is to ensure presentation of accurate information and clearly falls within the remit of consumer protection. 
The issue of whether consumer protection is needed with direct-to-consumer tests is a whole other debate with several angles to approach the problem from.

The first is that consumers should be sold something that's actually useful to them.  If the product is presented as something with a use that's beyond satisfying mere curiosity, such as poster-size images of 'personal genetic profiles' of blown up electrophoresis gel or karyotypes, it has to be something they can use.  If the product doesn't fulfill consumer expectations, most will conclude that they've wasted their money on a high tech test of limited use to them. 

Some of these people might fall in the group that consults professional geneticists to discuss their results, and a set of these could end up disappointed that their $299 direct-to-consumer genetic test ended up costing $2,999 once they finish up with genetic counseling. 

The second issue is that even if the reports have a use, the quality of conclusions should actually be actionable. 

Knowing that you have an allele carried by 2% of the population that's related to metabolism of sugar is one thing, but knowing that it significantly increases your personal risk of developing diabetes is better.  That's actionable.  Knowing that you're a carrier of dystrophin mutations and have a risk of bearing children with Duchenne Muscular Dystrophy is actionable. 

Providing results that are merely interesting, but largely academic in nature, in a direct-to-consumer test should be limited despite it being possible to spit everything out of a database.  There's the risk of information overload, especially among consumers, and after a certain point the additional information decreases the value of the product.  Pigging out at a buffet makes you ill, not satisfied.

The third point for regulating emerging genetic tests concerns the need to respect and maintain public confidence.

The expectations for most new technology are high.  If providers rush in to sell direct-to-consumer tests before the results and reports are mature enough, they'll create the perception that the field is a pile of 21st century snake oil.  If consumer tests don't deliver in this start up period and fall short of expectations, it'll be a while for the marketplace to forget and make it much harder to get any support for later genetic tests that actually provide great value.

In the end, regulation of new direct-to-consumer genetic tests is needed but shouldn't become a sledgehammer on innovation, but something that helps new technology become products of high quality that it should rightfully be.

Friday, August 16, 2013

Genome Liberty: Predicting Your Drug Response Using Genomics

Susan Young, at MIT Technology Review, describes how a start up is developing a genetic test to predict your personal response to drugs:
Consumers who get the Genome Liberty drug response test would send a sample of saliva to the company’s lab. The company scans the genome for DNA variations in 11 liver enzyme genes, which are a subset of the dozens of genes encoding enzymes for drug metabolism. Enzymes in the liver process drugs and can either deactivate or activate drugs, depending on the compound. Different people carry different versions or amounts of many of these enzymes, which can affect how they respond to drugs. Some patients may process a drug more quickly, more slowly, or perhaps not at all.
Genome Liberty says that variants in those 11 enzymes can affect the activity of nearly 80 drugs in the body. 
Tests like these are nice, simple, and possibly very inexpensive approaches to help prescribe drugs.  I suspect Genome Liberty is basing their technology on something straightforward like a multiplex PCR to amplify relevant sites withing each enzyme's gene, followed by pooling of samples from many individuals and some basic sequencing.  The cost of producing the data on each patient/customer would be a fraction of their $99 price target, leaving a nice profit margin for Genome Liberty.

Now while spending $99 may not be cost-efficient to optimize a dose of Prilosec, an antacid which can cost about 42 cents per day, if you have a prescription for something like Abilify, an anti-depressant that can cost thousands of dollars per year (even in Canada), sequencing based tests like Genome Liberty's can save a lot of money for a health system or individuals, if they're paying for drugs privately.

The biggest potential customer for tests like this are still organized health systems or insurance companies.  Particularly insurance companies, who I think would pay for pharmacogenomic tests like this if it reduces the number of prescriptions given to people that simply can't metabolize the drugs too well, if at all.

The big obstacle facing the company, however, still seems to be coming from the FDA:
It is still not clear whether genetic tests sold directly to consumers will come under regulatory scrutiny. In 2010, the FDA warned 23andMe and other consumer genetics companies that their services amount to medical devices and thus need regulatory approval. But since then, the U.S. government has not come up with clear rules for these companies.
Deciding whether these tests are medical devices would calm concerns among people developing new technologies, particularly among funders and investors.  At this point, I think it's likely that genetic tests will eventually be considered 'medical devices', as both the US and Canada define procedures used to diagnose diseases or abnormal physical conditions as such.

On the other hand, in situation involving pharmacogenomics, like Genome Liberty is in, it could be argued that tests don't actually diagnose an 'abnormal physical condition', as rates of drug metabolism can cover a wide range even among normal individuals.  But that distinction is a fine one to make, one I think is better left for lawyers, attorneys, and others legally inclined to expand upon.

Like it or not, genetic tests like this are going to be as routine as ordering a blood test is today.  Jeffery Rosenfeld, a co-founder, closes with this advice:
Physicians don’t always trust the results of direct-to-consumer tests and may not have clear medical guidelines for how to use it. But the connections between the liver enzyme variants and drug response are well-supported, says Rosenfeld. If a doctor doesn’t want to accept these results, he says, then “find a different doctor.”

Tuesday, August 13, 2013

Breaking Away from Powerpoint

At BitesizedBio, Vicki Doronina writes:
Powerpoint, and it’s free relatives, is that it limits your presentation style and makes your over-reliant on visual aids.
So what about taking it further and breaking away from the linear, prescribed talks that Powerpoint and the free alternatives tend to facilitate? This can be useful as it stops you from getting into a rut and helps you to explore other ways in which you can talk about your results that might be more engaging for your audience.
The best software for this that I have seen is Prezi. This is a free online software that allows you to organise your presentations in a non-linear fashion. Some people compare its organisation with poster, where you can see the whole picture at the same time and zoom into different parts. The site itself says that “a Prezi” can be a visual metaphor for the process and/or structure of your idea.
I haven't used Prezi, as it still doesn't free people from the tendency to pack a presentation full of data.  It would work extremely well for something like the Death and Taxes Poster for 2014 if you want to walk through the US Budget.  Nevertheless, someday I'll give Prezi a try.

One suggestion that I've started making is to escape from the "One Minute, One Slide" rule of thumb, which I've found generally doesn't work for most scientific presentations, unless you present data bit by bit.  Two to three minutes per slide gives you enough time to get your message through to your audience, which is what you want to do in the first place, right?

Monday, August 12, 2013

Tony Pawson: Last Thoughts on Science Policy

Most of you have probably heard that Tony Pawson, discoverer of the SH2 protein-protein interaction domain and mentor of many great biomedical researchers, passed away suddenly last week.  I never took advantage of the opportunity to meet him, but remember his talks as influential, information-packed, and motivating to all.

He also had a keen mind for seeing how the science's undercurrents have changed over the course of his career.  At Macleans, Paul Wells emphasized Pawson's views on science policy in Canada, quoting him thusly:
The ability of Canadian scientists to dream and make the kind of truly innovative discoveries that lead to awards is increasingly compromised by the declining levels of funding based primarily on excellence and vision, and the rise of support for so-called translational and targeted research, which is too frequently of questionable value. Canadian funding bodies are increasingly putting as much weight on non-scientific factors such as socioeconomic benefits as they are on scientific quality in deciding who to support.
They are entirely at liberty to do so, but they must realize that this strategy compromises innovation of the sort that will change the world. The funding that allowed me as a young assistant professor to do meaningful science 25 years ago is less readily available today, and without it we will not win more prizes, regardless of the volume of nominations.
The changes last 25 years have opened a niche for foundations to funding basic research.   Science is a very competitive environment and the attention given to basic and commercializable research comes in two very distinct styles, and along with Pawson's concerns for basic research, even industrial folds are increasingly vocal of the erosion of the basic research pipeline.

Tuesday, August 6, 2013

Cochrane's Aphorism for Scientists

I just recently heard Cochrane's Aphorism:
Before ordering a test, decide what you will do if it is (1) positive or (2) negative. If both answers are the same, don't take the test.
The origin is probably from clinical practice, but it seems apt to paraphrase it for scientists' use:
Before running an experiment, decide what you will do if it is (1) positive or (2) negative. If both answers are the same, don't do the experiment.
Much time, money, and distress would be saved in any field of research if only every PhD was taught this during their first month of work.

Friday, August 2, 2013

Eight World Innovation Clusters Ranked

The MIT Technology Review has ranked eight Innovation Clusters around the world.  One of the newest: Paris-Saclay, with about $3.25 billion in government backing.  Construction began this year.

Go big or go home.

Thursday, August 1, 2013

Awesome Animation of Earth's Seasonal Variation

If you've ever wondered what the rhythm of the seasons looks like from space, here's your chance.
Michigan programmer John Nelson created an absolutely mesmerizing animated video designed to provide a view of our planet “breathing” over the course of one year.
The foundation of the video is pictures from NASA’s “Blue Marble Next Generation” project, which takes a picture of the Earth each month.
The effect is just plain aweseome.  In his words:
“I can track one location throughout a year to compare the annual push and pull of snow and plant life there, while in my periphery I see the oscillating wave of life advancing and retreating, advancing and retreating. And I’m reassured by it.”
Just another great example of presentation style taking relatively simple data and taking it to the next level.

Micromanagement in Federally Funded Research in Canada Reaches Unheard Of Lows

A quip from the CBC on a predicted drop in Federally funded research in Canada for 2013-2014 comes with a bonus stock photo of Canadian Prime Minister Stephen Harper visiting a lab.

He's probably just visiting, but in light of previous complaints of micromanagement by the PMO, could point to a trend towards doing the same in labs across the country.

Stephen Harper personally ensures that science is being done right. (Source: CBC)