Wednesday, August 2, 2017

Backyard science: Aphids under a USB microscope

Just some random summertime backyard science here. The linden tree near my home is covered in aphids, so I took a few pictures with this USB microscope.

Can anyone confirm whether this cluster of eggs belongs to aphids or something else altogether?

Tuesday, August 1, 2017

The FDA, 23andMe, and selling less for more

Back in 2013 and at the beginning of the FDA vs 23andMe story, I wrote this:
Personal genetics will elicit a huge change in medicine, [but] actually doing that requires knowledge of how a person's genetics impacts their health.  For many genes, like BRCA1/2 which are the oft cited breast cancer risk markers in the media, the links between genetics and disease are established.

But for many others, a lot of work still needs to be done: clinical trials based on experiments need to be designed; experiments based on hypotheses need to be run; hypotheses flowing from data analyses need to be identified; and analyses need data sets to start from.  Fuzzy data is better than no data at all.

I expect that's where 23andMe is going to graciously come into play.  If the FDA actually winds up killing the company's main data collection channel, the company already has about 400,000 people genotyped and has built up enough traction to run for a while without having to sell another $99 test for a long time.

It took four years, but as of a few months ago, the FDA has reversed it's decision against 23andMe allowing the company to resume DTC marketing. The catch, however, is that the company will now offer tests for 36 conditions (instead of 254) for $199 (instead of $99). Gizmodo correctly points out that though this means that people are getting "less info for more money" (more on that in another post). Technically, that may be true, but if that information is more accurate or at least less irrelevant it counts for some of that value.

Monday, April 17, 2017

PCR's Next Wave

On BioTechniques:
PCR amplification of DNA is arguably one of the most important technology developments in the history of molecular biology, because without PCR, many commonly employed lab techniques would not be possible. It’s both robust and simple. Yet, there are still parts of the method that could be further refined—especially when it comes to sensitivity and error rate.

For researchers working in the cancer field, locating extremely rare variants from complicated mixtures of DNA, proteins, RNA, and other biomolecules is often a daily task. But identifying ultra-rare variants of these molecules presents a huge challenge—it’s like finding a needle in a haystack, a process that can be confounded further by PCR errors. This is another area where developers are finding fertile ground for improving PCR methods and protocols.

Anders Stahlberg from the University of Gothenburg in Sweden, along with colleagues from Boston University School of Medicine and the Ontario Institute for Cancer Research, recently described a new multiplex PCR-based barcoding strategy that can be used for the detection of ultra-rare mutations by next-generation sequencing.
This article is a nice review of the previously described SiMSenSeq paper.

Wednesday, February 22, 2017

Do European Research Council applications foreshadow what may happen with the NIH?

It looks like Brexit is having repercussions on UK science funding from the European Research Council, according to David Lomas:
Applications to EU funds are usually made by consortiums of researchers from a variety of EU countries. Britain has a strong track record of taking the lead in these groupings. 

Britain also has a leading position in applications for individual research grants made under the European Research Council, the ERC. 

“Previously having a British member would help you in your application to get funding … Now you are less than an asset, so we have had academics removed from grant applications,” Lomas said.
One has to wonder whether US funded research organisations, like the NIH, will be forced to "Buy American" in the future and prevent spending on research activities in foreign countries.

I get the whole populist sentiment around trade deficits and job losses, etc., but the fact of the matter is that the US can't just cut out part of international trade and partnerships without impacting the productivity of US-based initiatives. For example, groups in Toronto can offer US scientists many different kinds of scientific services on a fee-for-service or collaborative basis on very competitive terms, mostly due to the favourable USD/CAD exchange rate.

So why cut out that part of the value chain?