Tuesday, February 26, 2013

Exome sequencing wins where traditional medical tests fail

This year's Advances in Genome Biology and Technology conference was densely packed with clinical applications of high-throughput sequencing, with the theme becoming a clear winner throughout the few days of AGBT.  Memorable talks by Christine Eng and Stephen Kingsmore last Thursday provided a high level overview regarding the state of the field.

Saturday, February 23, 2013

Illumina Unveils Little About Long Read Technology at AGBT

This year's Advances in Genome Biology and Technology conference in Marco Island, Florida is as packed with exciting talks and announcements as last year's, though notably with less hype about new sequencing technologies.

One eagerly awaited talk entitled "10 kilobase Reads on a HiSeq" was given by Geoff Smith, from Illumina, which covered a few updates to their product pipelines, but mainly focused on their recently acquired technology from Moleculo that’s intended to enable read lengths of 10 kilobases on standard Illumina instruments.

Monday, February 18, 2013

Beijing Genomics Institute's attempt to find genetic basis of IQ

From an essay by Gautam Naik, in The Wall Street Journal:
At the Hong Kong facility, more than 100 powerful gene-sequencing machines are deciphering about 2,200 DNA samples, reading off their 3.2 billion chemical base pairs one letter at a time. These are no ordinary DNA samples. Most come from some of America's brightest people—extreme outliers in the intelligence sweepstakes.
The majority of the DNA samples come from people with IQs of 160 or higher. By comparison, average IQ in any population is set at 100. The average Nobel laureate registers at around 145. Only one in every 30,000 people is as smart as most of the participants in the Hong Kong project—and finding them was a quest of its own.
"People have chosen to ignore the genetics of intelligence for a long time," said Mr. Zhao, who hopes to publish his team's initial findings this summer. 
Besides answering a very interesting question about human intelligence that's also controversial in the West (as the article concedes), this whole project is an immense example of the juggernaut that is the Beijing Genomics Institute.  While today's research projects routinely examine a few hundred genomes, if that, the "BGI Intelligence Project" aims to sequence and analyze several thousand.

What to do with a science PHD?

Labguru sent a newsletter today, containing an overview of career possibilities for science PhDs.  I had no idea the numbers changed so dramatically over the last forty years:
“in 1973, 55 percent of PhD recipients had tenure-track positions within six years of earning their PhDs. In 2006, merely 15 percent of recent graduates found themselves in this position.”
Whether the cause of this is employer driven 'academic inflation' or research driven reliance on 'graduate student and postdoc labor for publications and output', as Labguru suggests, is a matter for another day of finger pointing.  For now, Labguru's intended audience seems strange to me:
If you are reading this while you’re on your second, third or even fourth postdoc, you’re not alone, and are far from the minority.
I don't think I know anyone that's on a fourth postdoc.  If there are readers on their fourth postdoc, please tell me your story, regardless of whether it's good or bad!

Friday, February 15, 2013

Academic Research, Clinical Trials and Insider Trading

I wrote a short article for Signals Blog (a blog led by Canadian Stem Cell Network and the Centre for Commercialization of Regenerative Medicine) about recent cases of academic research and insider trading, which has become a topic of serious consideration for research institutions:
The current bellwether of leaked research information involves Sidney Gilman, based at the University of Michigan, and Mathew Martoma, a trader at SAC Capital. They met through the Gerson Lehrman Group, ... one of many ‘expert network’ companies hiring researchers for as much as $1,000 an hour. In 2008, Gilman passed information on an Alzheimer’s drug clinical trial involving Bapineuzumab to Martoma, who was arrested by the FBI in late November 2012.
The magnitude of this particular case is remarkable; US$276 million gained for the hedge fund involved and two careers ended in a spectacular flash.

Read High stakes: Unpublished research brings big profits and huge losses – to insider traders at Signals.

Wednesday, February 13, 2013

Will Public Funds Ever Follow the "People Not Projects" Approach

Barack Obama glossed on science funding, in yesterday’s US State of the Union address. He didn’t quite explain why research funding shouldn’t be cut, but I’ll explain in a second. First, Obama pitches the vision of economic returns:
Now, if we want to make the best products, we also have -- have to invest in the best ideas. Every dollar we invested to map the human genome returned $140 to our economy. Every dollar.

Sunday, February 10, 2013

Marseille Police Facing $1.3 Million Price Tag for Three Genomes

The city of Marseille, France, is considering a very peculiar scientific conundrum: one of two identical twins is a suspect facing sexual assault charges, but standard DNA testing doesn't have the resolution to differentiate between the two:
One expert told the French newspaper La Provence: "For a normal analysis, we would compare 400 base pairs [of nucleotides] which make up DNA."  In the case of identical twins, he added, "We would be looking at billions."
It's implicitly assumed that three whole genomes worth of sequencing (two for the twins, one for the evidence) are being considered to resolve which twin was the culprit.  However, the price tag is exorbitant:
Police have been told it would cost upwards of 1m euros (£850,000, or CAD$1.34 million) to conduct an ultra-sophisticated genetic test that would be able to tell one set of the twins' DNA from the other.
I have no idea where they're getting their cost estimates from.  Some lab in France is going to walk away with a tidy sum as whole genomes do not cost that much to run (I'll be generous and say that they're under $10 thousand each in Toronto; and at reasonable depths of coverage the cost can be half that).  The only explanation I can think of is that they're including the cost of an Illumina HiSeq in the price, maybe even two in case one breaks down during the investigation.

Though I'm pretty convinced that the Marseille police department needs the genome sequences on these twins to base their evidence, it's also known that chip based analysis of monozygotic twins can identify hundreds of Single Nucleotide Polymorphisms differing between the pair.  One study identified 477 different variations among 500k examined.

That's not to say that whole genome sequencing will make their job any easier, despite the higher resolution and price tag.  One recent report sequenced three sets of twins, but failed to find any gene copy number variations or SNPs that differed between paired siblings, while another report examined the genomes of nineteen sets of twins for copy number variations using chip-based technology did.

The Affymetrix 500k SNP chip arrays, like the ones used in the first study, cost a few hundred dollars to run.  So again, I'm puzzled as to why the police in Marseille received such a wild estimate for this project: the data analysis will surely cost more than the sequencing, but $1 million dollars?  No way!

Friday, February 8, 2013

Big Data, insurance companies, and how both can improve people's health

Kat McGowan, interviewing Colin Hill, of GNS Healthcare:
Defining personalized medicine only in the context of genomics for drug discovery, patient stratification, and biomarker discovery is thinking too small, says Hill, a former computational physicist who founded GNS Healthcare, a health data analytics company, in 2000. We need to consider the whole universe of clinical information generated by randomized clinical trials, claims, electronic health care records, payers and providers—what’s being called the “data exhaust” of the digital healthcare era.
You have to read this article - it provides a small snippet of what big data sets can do to improve people's health.

Wednesday, February 6, 2013

Water fluoridation may have no effect on children's IQ

Online new sites erupted in criticism and some fanfare to the City of Windsor's decision to stop fluoridating their drinking water, with Jonathan Kay at the National Post staunchly against the move.  There are ongoing arguments against fluoridation of water, like those summarized by Dr. Gifford-Jones at the Toronto Sun last fall, but since the water fluoridation debate has evolved over at least 35 years, I can't see it going away any time soon.

Tuesday, February 5, 2013

Biorefinement of soluble gold

Andy Coghlan, at New Scientist:
Delftia acidovorans ... has evolved a chemical that detoxifies gold ions by turning them into harmless gold nanoparticles. These accumulate safely outside the bacterial cells.
Very cool!  One of the discoverers, Nathan Magarvey, of McMaster University in Canada thinks that the protein responsible could be used to create sensors capable of detecting gold ions in groundwater.  It wouldn't exactly extract much gold but it could provide an inexpensive method of prospecting by taking water samples. 

How to screw up chemo delivery

Tom Blackwell, for the National Post, reports on a study revealing mistakes in the chemo delivery to patients:
Problems included attaching the wrong infusion pump to a patient, incorrectly calculating the volume of drugs, and “numerous cases” of the pump being clamped shut so the patient absorbed none of the medication. ... The team highlighted three serious flaws: The lack of a second worker in some pharmacies to check whether a chemotherapy solution had been properly combined; work surfaces where several bags of chemicals were being prepared at one time, raising the risk of mix-ups; and failing to keep a label bearing the patient’s name attached to the infusion bag at all times, increasing the chances of giving patients the wrong drugs. 

I can't believe that such simple mistakes are that common, so I wanted to see what the original authors said.  Unfortunately, the original article (found here) can't be accessed by the general public and you need institutional access to read it, but the actual report makes a valuable conclusion:

Friday, February 1, 2013

Terahertz radiation affects stem cells?

Alexandrov et al. in Nature Scientific Reports show how terahertz radiation can affect mesenchymal stem cells and change the expression of genes:
Our microarray survey and RT-PCR experiments demonstrate that prolonged broadband THz irradiation drives mMSCs toward differentiation, while 2-hour irradiation (regardless of THz sources) affects genes transcriptionally active in pluripotent stem cells. 
Besides being a very weird (but cool) observation, I imagine that this would lead to the ability to specifically change gene activities using radio frequency devices in some cases and avoid pharmaceuticals.  That would be very, very cool indeed.

Making influenza more efficient: flu virus engineering resumes

Declan Butler, at Nature, writes:
The move follows a large international workshop convened on 17–18 December by the US National Institutes of Health in Bethesda, Maryland, to discuss ‘gain-of-function research’ — that intended to increase the transmissibility, host range or virulence — in H5N1 viruses, and the development of US rules for stricter oversight of research in this area. The proposed rules require an assessment of, for example, whether the scientific aims of such studies could be addressed using alternative, less-risky approaches, and whether biosafety and biosecurity risks can be adequately mitigated. They are expected to enter into force soon, allowing scientists working in the United States or on US-funded grants to restart such research.
The groups that published the original research have outlined a suite of possible follow-up experiments, including a search for other combinations of mutations that would allow H5N1 to transmit between mammals — which could answer basic-science questions and, they argue, aid efforts to watch for dangerous mutations in the wild.
Two papers discussing influenza engineering were published last summer; one in Nature and one in Science.  At first glance, you'd assume they're another case of competitive research completing the same project almost simultaenously (a very common happenstance), but the Science team from Erasmus writes that "the viruses that caused the major pandemics of the past century emerged upon reassortment (that is, genetic mixing) of animal and human influenza viruses".  They continue to explain that viruses generated through reassortment events don't work very well in ferret models, acknowledged that limitation, and continued to explore other genetic alterations that increase ferret-to-ferret transmission. On the other hand, the Nature team didn't acknowledge any limitation of viruses created by genetic reassortment, and actually base their study around viruses generated through that method, which according to the Erasmus team shouldn't work very well.

If anyone can explain this better to me, please let me know.