Just days after a lab mix-up at personal genomics company 23andMe, the US Food and Drug Administration is demanding federal regulation for genetic tests marketed directly to consumers. According to the agency, the tests are medical devices, and should be evaluated as such.

The FDA sent letters to five companies on Thursday, saying the tests must go through an approval process before they go on the market. 23andMe and two other companies selling the tests – Navigenics and DeCode Genetics — received the letters, as did the sequencing company Illumina, which sells the gene arrays some companies use in their products, and Knome, which offers consumers whole genome sequencing. Company responses have been variable, the New York Times notes, with 23adnMe saying it disagreed with the FDA’s decision but DeCode saying it welcomed regulation.

You can read the letters from the FDA here.

The FDA isn’t alone in getting antsy about personal genomics. The agency’s demand last month that the Walgreens pharmacy chain halt plans to sell a home genetic test spurred Congress to embark on an investigation of the direct-to-consumer personal genomics industry.

The San Francisco Chronicle had a thoughtful piece yesterday on the industry as a whole, and thoughts surrounding regulations – check it out here.

After a genome-wide analysis last week reported that Jews worldwide have significant genetic commonalities, a second study published today comes to a similar conclusion. But will the new evidence put to rest a debate that has lasted more than a century?

Tradition has it that Jews around the world are one people with shared religious and cultural practices. Several studies from the last decade that have looked closely at genetics have pointed to a common history based on shared genetics between different Jewish populations. But many historians doubt that genes could be shared, noting that there was so much mixing between Jews and neighboring populations – non-Jews converting to Judaism in large numbers around or before the time of Christ, or Jews being sold as slaves and sent off to far-off lands, then freed there – that they should have come to genetically blend into the geographical population where they ended up.

Catherine Hezser, who studies ancient Judaism at the University of London’s School of Oriental and African Studies and wasn’t involved in the research, notes that for this reason, she does not “believe in any of the genetic theories, which are usually propagated for non-scientific, ideological reasons”. But the researchers say their fine-grained analysis of 14 Jewish populations, published in Nature today, could provide some answers.

The groups examined in the study included not just major communities such as Ashkenazi Jews of Eastern Europe and Sephardim from Bulgaria and Turkey, but also several that are much smaller and, like the Bukharan Jews of Central Asia and Jews from India, Ethiopia and Yemen, are often referred to as the “lost tribes” of Israel. The study mainly examined single-letter variations – called single nucleotide polymorphisms (SNPs) – in individuals within each group, compared to individuals in surrounding areas.

Much like the analysis published last week, the study showed that all of the Jewish communities share some common genetic features, and for the most part, the Jewish groups are more similar to each other than to the non-Jews in the same regions.

“These two studies are the first pair of genome-wide studies of SNP variations in collections of multiple Jewish populations,” says Noah Rosenberg, a population geneticist at the University of Michigan who was not involved in either study. The strength of the Nature paper, he says, is that it compared the Jewish groups to an unprecedentedly broad array of non-Jewish groups, making the comparison – and the genetic ties identified – especially robust.

Continue reading →

Stanford University School of Medicine announced earlier this week that it would be offering a new class on genetic testing, in which students would be able to get their own DNA tested by personal genomics companies and then learn to interpret the results.

Students would first learn about the legal and ethical issues in DNA testing, and then would have the option of getting an analysis of their own DNA from personal genomics companies 23andMe or Navigenics – and paying $99 for the privilege – or working with publicly available genomic data instead to learn how to interpret the results of such tests.

One wonders whether classes like these will soon become the norm for doctors in training. As the importance of genomics in health care grows, there has been a call to better train medical professionals to interpret such information for patients – and what better way to do so than with their own DNA? According to Stanford’s announcement, the course, to be offered this summer as an elective, is the first of its kind. Genomeweb points out, however, that the University of Pennsylvania recently offered a similar course. And last year, Harvard Medical School set up a program in which medical residents could also receive training on interpreting results of personal genomics tests by offering up their own DNA for study.

Stanford says its course was proposed by a student, and that a task force spent a year considering the risks and instituting safeguards to protect students from feeling coerced into participating. Meanwhile, last month, vociferous debate ensued when the University of California, Berkeley, said it would offer incoming freshmen and transfer students a similar learning opportunity. This summer, the undergraduates will receive a cotton swab in the mail with their welcome packets, which they can choose to send in to test for three genes metabolizing alcohol, lactose and folates. A Berkeley-based organization, the Center for Genetics and Society, immediately called for the program’s suspension, saying the tests weren’t yet well regulated and shouldn’t be interpreted without a medical professional.

Indeed, sometimes even a medical professional isn’t enough. Speaking to a group of science writers earlier this week, Francis Collins, director of the US National Institutes of Health, said he had sent in a saliva sample under a pseudonym to three different personal genomics companies – DecodeMe, 23andMe and Navigenics. While reports from all three companies said he was at high risk for diabetes, he got back three different risk levels – low, average, and higher than average – for prostate cancer.

Due to errors in the lab, as many as 96 customers of consumer genomics company 23andMe “may have received and viewed data that was not their own”, the company informed customers in an announcement on its website (password protected) last Friday.

The announcement says that as soon as the company became aware of the problem, “we immediately identified all customers potentially affected, notified them of the problem and removed the data from their accounts”.

“We are currently putting additional procedures in place that will add an extra layer of safeguards to help assure that similar incidents do not occur in the future,” the statement goes on to say. “We are deliberating on a process that would include removing manual steps at the lab, completely automating the sample analyses, and implementing further checks of the data before it gets loaded into customer accounts.”

Some of the comments posted to the announcement – a few of which were written by customers affected by the mistake — praise the company’s transparency, while others suggest the company failed to respond to the situation in a timely fashion.

Daniel MacArthur at the blog Genetic Future, who first reported the mix-up this morning, points out that while mistakes can happen anywhere, the industry needs to be particularly careful to maintain its integrity and “avoid any appearance of being amateurish”.

In this case, the damage was relatively minor and quickly rectified, says Sharon Terry, president and CEO of the Genetic Alliance, a non-profit genetics advocacy group based in Washington DC. But she believes the area of genetic testing overall requires “multiple levels of oversight and assurance”, pointing to a recent case in which a genetic test for breast cancer made by a company called Mammaprint had over-reported patients’ metastasis risk by 15% for about 6 months.

In addition to other measures, “there have to be transparent registries, like the Genetic Testing Registry at NIH, and I believe it and other mechanisms should be mandatory to reduce the chaos in the system”, she writes in an email to Nature.

MacArthur also notes that it’s not just regulators who need to be vigilant for errors, but customers themselves, who should be engaging with the information such companies provide rather than passively receiving it. As he points out in his post, “…it is noteworthy that in this case the sample problems seem to have been detected by customers rather than the company”.

The vertebrate circulatory system, to sum it up glibly and way too simply, is a heart pumping blood. But how and when in an organism’s development does the flow begin?

A series of movies that capture the process in developing zebrafish show that it requires a ‘go’ signal. Hundreds of blood cells wait in the equivalent of a vascular starting gate, tethered in place to the blood vessel walls. Then, all at once, they are set free by an enzyme called ADAM8, which acts as a ‘molecular scissors,’ Atsuko Sehara-Fujisawa of Kyoto University and her colleagues describes in a paper published online in Current Biology yesterday. Without the enzyme, blood cells stagnate in the starting gate. (Press release here.)

The researchers don’t yet know whether a similar mechanism works in other organisms, but they suggest three possibilities for why blood cells wouldn’t be released to circulate in the body one by one, as they come off the production line. First, an active signal would provide control for which cells go and which cells wait. Or, it might help the developing organism to make sure vessels are fully formed before they fill with blood. Finally, there are some hints, the researchers say, that blood cells need plasma to flow first in order to fully mature.

In timelapse photos from the researchers’ videos, below, you can see red blood cells accumulating a blood vessel, then beginning to freely flow. (The action takes place in the PCV, or posterior cardinal vein. In the first image, taken 24 hours postfertilization (hpf) the lumen of the dorsal aorta (DA) contains few erythrocytes. In the second and third image, the vessel is more fully formed and begins to fill with blood cells.The fourth image shows the onset of circulation, with the process in full force in the final image.)

Sequenom, a San Diego based biotech company developing prenatal genetic tests, is in hot water again – this time with the US Securities and Exchange Commission (SEC).

Almost exactly a year ago, executives at Sequenom admitted that four employees had “mishandled” data supporting the company’s noninvasive prenatal test for Down syndrome, and that the test would have to be validated anew. Now, the federal regulator has charged the company’s former head of research, with lying about the accuracy of the test in at least three public presentations to investors and analysts, Reuters reports.

In its complaint, filed yesterday, the SEC charges that Elizabeth Dragon, then Sequenom’s vice president of research and development, publically stated on three occasions between June 2008 and January 2009 that the test was nearly 100% accurate, while knowing this was not the case. The complaint goes on to say that studies on the test’s validity were not blinded, as Dragan said they were, and that she revealed the identity of samples to other employees in order to skew the results in favor of its accuracy.

According to Reuters, the company still hopes to develop the test, and said last month it plans to launch it by the end of next year.

A scientific misconduct investigation at a research laboratory at the Mayo Clinic in Rochester, Minnesota, has derailed a promising research project aimed at developing a potent immune system booster for cancer immunotherapy.

The investigation, conducted earlier this year, has also led to the retraction of between 15 and 17 papers published between 2002 and 2009 in nine journals and to the termination of a scientist who had worked in the lab for almost a decade, according to the Mayo Clinic. The reseracher, Suresh Radhakrishnan, says he has been wrongly dismissed.

Researchers in the lab of Larry Pease eight years ago identified an antibody in a patient’s serum which appeared to stimulate the function of dendritic cells in the immune system, potentiating immune response. The group had developed a cancer vaccine strategy based on the research, and the project had become a main focus of the lab’s work.

According to Robert Nellis, a Mayo Clinic spokesperson, researchers in Pease’s lab began to get different findings in working with the antibody than those the group had previously reported in published work. They tried unsuccessfully to replicate those experiments. Pease then turned to the Mayo Clinic authorities, accusing Suresh Radhakrishnan, a researcher in the group, of tampering with their attempts to validate the past work. An investigation launched by the institution found Radhakrishnan guilty of scientific misconduct. The institution fired Radhakrishnan, and the lab decided to retract all of the published work containing data that could not be verified, Nellis says.

Some of the retractions are still in progress; six retraction notices were posted in the June issue of the Journal of Immunology. [Hat tip: DrugMonkey]

“The laboratory has moved on and refocused its efforts along other lines,” Nellis says, and plans to commercialize the work have been halted.

Continue reading →

Just when you thought you knew everything there is to know about Neanderthals, Svante Paabo, the Swedish researcher who sequenced the Neanderthal genome, throws in another sequenced ancient hominin.

Speaking at the Biology of Genomes meeting in Cold Spring Harbor last night, Paabo revealed that his group has now sequenced nuclear DNA from a bone fragment found in Denisova Cave in Siberia, thought to belong to an extinct, ancient hominin that branched off from the human family tree about a million years ago. The group reported sequencing the mystery hominin’s mitochondrial DNA in March, but couldn’t draw firm conclusions about its relationship with humans and Neanderthals without its nuclear genome.

The major form of damage that occurs to ancient DNA involves the conversion of the base cytosine to another base, uracil. The trick to sequence it was to basically remove the uracil, cutting the DNA at that spot, then sequence the many short snippets thus produced and line them up.

Comparing the sequence to Neanderthal and human sequences revealed that the Denisova hominin is quite distant from humans, sharing a closer common ancestry with Neanderthals. How far back that split occurred isn’t clear, Paabo says, but the Denisova hominin “did not take part in the gene flow from modern humans.” The hope, of course, is that this is just the start, and that other ancient hominin samples will yield sequences as well.

Much has been said about a mother’s bond with her young, but little is known about the biological glue that ties males to their offspring. Now Canadian scientists show that when male mice interact closely with their babies, their brains grow new neurons that help them recognize their pups weeks later, when they have grown into adults.

Male mice who stay with females during and just after pregnancy spend their time nuzzling in the nest. Those animals were able to recognize offspring after they grew up, but mice removed from the nest after the litter was born could not, Gloria Mak and Samuel Weiss of the University of Calgary in Alberta, Canada, found. Fathers that stayed with the new litter also showed a burst of new neurons in two regions of the brain linked to the ability to differentiate odors – the olfactory bulb and the dentate gyrus. What’s more, those neurons preferentially responded to the offspring’s smell.

In females, the hormone prolactin induces new neuron birth during pregnancy or after exposure to male hormones, so the researchers wondered whether the same hormone was at play with males too. Indeed, males lacking the prolactin receptor didn’t produce new neurons and were unable to recognize their offspring. When neurogenesis was induced with another hormone, however, recognition was restored. The study appeared online in Nature Neuroscience yesterday, and the researchers speculate that the same mechanism might work in humans, too.

Opponents of a much-contested set of stem-cell patents are claiming victory with a decision from the US Patent and Trademark Office (USPTO), which declared one of the three patents invalid.

The patents, held by the Wisconsin Alumni Research Foundation (WARF), cover methods for deriving and growing primate and human embryonic stem cells and are based on the work of James Thomson, a University of Wisconsin stem-cell researcher done in the 1990s. In 2006, a group led by the New York-based Public Patent Foundation led efforts to contest them, claiming that the research on which they were based was not novel enough to be patented, that the patents’ claims were too broad, and that they impeded stem cell research. In response, WARF narrowed the claims, and the USPTO upheld all three patents in 2008 (see our coverage here and here). While two of the decisions were final, one was eligible for appeal.

Last week’s reversal of that patent’s validity effectively invalidates the other two patents as well, says Daniel Ravicher, the executive director of the Public Patent Foundation. “The emperor has now been proven to have no clothes.” If WARF were to try to sue anyone with the two valid patents, he says, the decision negating the third one would serve as a precedent.

Continue reading →