holycrapscience! @holycrapscience-blog - Tumblr Blog

Posts

Abstract: This is a two part-post about a science piece on gene regulation that just appeared in the New Yorker. Today I give quotes from scientists criticizing that piece; tomorrow I’ll pres…

Do my fellow academics agree with his general critique of @newyorker‘s science coverage? Feels on pt to me.

#science #journalism #epigenetics

Advanced genetic tools could help boost crop yields and feed billions more people.

via Dave

Congrats to Dan and Elad on their paper coming out. I think these results are super surprising and very interesting.

#systems biology #pnas

Entirely predictable that Dan Longo and Jeffery Drazen’s editorial “Data Sharing” in the New England Journal of Medicine is not accessible to the public. In the spirit of data sharing, I’ve pasted their drivel below. When the editor in chief of the New England Journal thinks like this, it’s no surprise that they retract a larger fraction of papers than Nature, Science, Cell or the Lancet. http://iai.asm.org/content/79/10/3855.full

Jonathan Peele’s translation of this mess into English is on point: http://jonathanpeelle.net/blog/2016/1/22/translation-to-plain-english-of-selected-portions-of-longo-and-drazens-editorial-on-data-sharing

EDITORIAL

Data Sharing

Dan L. Longo, M.D., and Jeffrey M. Drazen, M.D.

N Engl J Med 2016; 374:276-277January 21, 2016DOI: 10.1056/NEJMe1516564

The aerial view of the concept of data sharing is beautiful. What could be better than having high-quality information carefully reexamined for the possibility that new nuggets of useful data are lying there, previously unseen? The potential for leveraging existing results for even more benefit pays appropriate increased tribute to the patients who put themselves at risk to generate the data. The moral imperative to honor their collective sacrifice is the trump card that takes this trick.

However, many of us who have actually conducted clinical research, managed clinical studies and data collection and analysis, and curated data sets have concerns about the details. The first concern is that someone not involved in the generation and collection of the data may not understand the choices made in defining the parameters. Special problems arise if data are to be combined from independent studies and considered comparable. How heterogeneous were the study populations? Were the eligibility criteria the same? Can it be assumed that the differences in study populations, data collection and analysis, and treatments, both protocol-specified and unspecified, can be ignored?

A second concern held by some is that a new class of research person will emerge — people who had nothing to do with the design and execution of the study but use another group’s data for their own ends, possibly stealing from the research productivity planned by the data gatherers, or even use the data to try to disprove what the original investigators had posited. There is concern among some front-line researchers that the system will be taken over by what some researchers have characterized as “research parasites.”

This issue of the Journal offers a product of data sharing that is exactly the opposite. The new investigators arrived on the scene with their own ideas and worked symbiotically, rather than parasitically, with the investigators holding the data, moving the field forward in a way that neither group could have done on its own. In this case, Dalerba and colleagues1 had a hypothesis that colon cancers arising from more primitive colon epithelial precursors might be more aggressive tumors at greater risk of relapse and might be more likely to benefit from adjuvant treatment. They found a gene whose expression appeared to correlate with the expression of genes that characterize more mature colon cancers on gene-expression arrays and whose product was reliably measurable in resected colon cancer specimens by immunohistochemistry. To assess the clinical value of this potential biomarker, they needed a sufficiently large group of patients whose archived tissues could be used to assess biomarker expression and who had been treated in relatively homogeneous way.

They proposed a collaboration with the National Surgical Adjuvant Breast and Bowel Project (NSABP) cooperative group, a research consortium funded by the National Cancer Institute that has conducted seminal research in the treatment of breast and bowel cancer for the past 50 years. The NSABP provided access to tissue and to clinical trial results on an individual patient basis. This symbiotic collaboration found that a small proportion (4%) of colon cancers did not express the biomarker and that the survival of patients with those tumors was poorer than that of patients whose tumors expressed the biomarker. Furthermore, when the effect of adjuvant chemotherapy was assessed, nearly all the benefit from adjuvant treatment was within the biomarker-negative group, the patients with the most primitive tumors. The findings have generated a new hypothesis that is now ready for testing in a prospective randomized clinical trial.

If the hypothesis that nearly all the benefit from adjuvant chemotherapy is in the biomarker-negative group is confirmed, over 90% of patients with stage II colon cancer will be reassured that avoiding the unpleasantness of standard adjuvant therapy is unlikely to affect their outcome adversely. No one expected that.

How would data sharing work best? We think it should happen symbiotically, not parasitically. Start with a novel idea, one that is not an obvious extension of the reported work. Second, identify potential collaborators whose collected data may be useful in assessing the hypothesis and propose a collaboration. Third, work together to test the new hypothesis. Fourth, report the new findings with relevant coauthorship to acknowledge both the group that proposed the new idea and the investigative group that accrued the data that allowed it to be tested. What is learned may be beautiful even when seen from close up.

#nejm #open access

How much of published scientific research is false? Scientists are trying to figure it out.

@planetmoney covers a very important issue in the empirical sciences - that a large fraction of scientific studies do not replicate when we try them again. My feeling is that we should start funding replication efforts more thoroughly and lower the acceptable P-value threshold from 0.05 to 0.001.

#science #statistics

Francis Crick exceeded the brilliance of his famous double helix model by this magnanimous act:

via Nikolai Slavov

Beautiful recollection by Sydney Brenner of Crick’s awesomely large-hearted response to being scooped by Marshall Nirenberg in the elucidation of the genetic code.

#history of science #menschlichkeit

In 1977 Carl Woese and George Fox published a 2.5-page paper1 in the Proceedings of the National Academy of Sciences. Their paper had no figures and only a single table, but it would ultimately set the tone for the next 35 years of research on bioinformatics and the evolution of life.

I interviewed George Fox on becoming a scientist and, along the way, the history of life on Earth.

Very cool paper that uses modeling methods from mechanical engineering to ask the question: is the morphology of Neanderthal bone specimens consistent with speech? The hyoid bone supports the tongue and, in humans, is shaped in such a way that allows for speech. Many non-human primates have different shaped hyoid bones, where the shape difference often accommodates their laryngeal air sacs (the air sacs might help apes produce loud and complex calls without hyperventilating). This paper shoes that a Neanderthal hyoid specimen from Kebara, Israel is shaped very similarly to a human hyoid and could support the mechanical forces and stresses that are associated with human-type speech.

#physiology #archaeology #modeling

It is usually assumed that modern language is a recent phenomenon, coinciding with the emergence of modern humans themselves. Many assume as well that this is the result of a single, sudden mutation giving rise to the full “modern package.” ...

This was a fascinating read. Turns out that there has been a long standing presumption that Neanderthals could not speak. According to this review, that assumption is not very well supported by genetic, archaeological and physiological evidence. For example, Neanderthals have the right skeletal morphology for speech and they have the right genes (at least for the few speech related genes that we know of). These guys are basically arguing that anthropocentrism is what kept the “Neanderthal’s were big dumb mutes” idea alive.

#human evolution #neanderthal #phylogeny #archaeology

Einstein’s theory of general relativity, explained using only the thousand most common words in English.

What do folks think of this? I like the premise of making important science accessible to a non-science audience, but I think the conceit actually made it harder to follow than a colloquial explanation.

#special relativity #einstein #xkcd

In his book “Fractals, Chaos, Power Laws,” the physicist Manfred Schroeder presented a breathtakingly simple proof of the Pythagorean theorem whose provenance he traced to Einstein. Schroeder wrote that the proof had been shown to him by a friend of his, the chemical physicist Shneior Lifson, of the Weizmann Institute, in Rehovot, Israel, who heard it from the physicist Ernst Straus, one of Einstein’s former assistants, who heard it from Einstein himself. Though we cannot be sure the following proof is Einstein’s, anyone who knows his work will recognize the lion by his claw.

#tradition symmetry

CRISPR technology has also made it possible to study complicated illnesses in an entirely new way. A few well-known disorders, such as Huntington’s disease and sickle-cell anemia, are caused by defects in a single gene. But most devastating illnesses, among them diabetes, autism, Alzheimer’s, and cancer, are almost always the result of a constantly shifting dynamic that can include hundreds of genes. The best way to understand those connections has been to test them in animal models, a process of trial and error that can take years. CRISPR promises to make that process easier, more accurate, and exponentially faster.

Thanks to @newyorker for an accessible description of CRISPR/Cas9 that mentions the most important feature of the technology: that it has and will greatly accelerate the pace at which we learn things about genetics and molecular biology of many diverse organisms. Designer babies and toucantelopes are just for show.

#crispr #genetics #molbio #popsci

I confess, my highschool biology teacher sold us the XYY = criminal line too.

#myths #karyotype

If you follow the headlines, your confidence in science may have taken a hit lately. Peer review? More like self-review. An investigation in November uncovered a scam in which researchers were rubber-stamping their own work, circumventing peer review at five high-profile publishers.

Via @__aston__: 538 on the (statistical) difficulties getting answers from the universe. Science is hard folks! Also, the P-hacking game is super fun and instructive.

I am a little disappointed that CHRISTIE ASCHWANDEN doesn’t discuss corrections for multiple hypothesis testing at all. The following paragraph was a perfectly excellent opportunity to bring this issue up:

Which political party is best for the economy seems like a pretty straightforward question. But as you saw, it’s much easier to get a result than it is to get an answer. The variables in the data sets you used to test your hypothesis had 1,800 possible combinations. Of these, 1,078 yielded a publishable p-value, but that doesn’t mean they showed that which party was in office had a strong effect on the economy. Most of them didn’t.

If we check all 1800 possible pairs, the Benferroni correction demands that we divide our P-value threshold by 1800 as well. This gives a threshold of 2.8e-5. You can also think of this as multiplying your P-value by 1800. Either way, you would find no significant results from P-hacking in this data if you Bonferroni correct the P-values. The Benferroni correction is the most stringent of the corrections for false discovery, but no matter which of the various sensible controls you used here it would solve the problem. If journals insisted that all P-values were adequately corrected for multiple hypotheses it would go a very very long way.

#science #statistics #reproducibility #538

Very surprising to see the “co-founder of Greenpeace,” Patrick Moore, give such full throated support for genetically modified foods. I was initially very happy to see this implicit repudiation of all the work that Greenpeace does to block the development and marketing of Golden Rice. But then it turns out that PragerU is doing a little deceptive marketing: Patrick Moore is no longer affiliated with Greenpeace and works as a corporate lobbyist. Is there any pure joy online? See some of Greenpeace’s anti-Golden Rice screed here: http://www.greenpeace.org/international/en/campaigns/agriculture/problem/genetic-engineering/Greenpeace-and-Golden-Rice/ Such a disgusting attempt to mislead by tugging on the twin heartstrings of naturalism and romance. Don’t forget how high the stakes are. Greenpeace is both distorting science and ignoring a substantial public health benefit.

#greenpeace #golden rice

websearch experiment

I made duckduckgo my default search engine on googlechrome for the last 2 weeks. While I appreciate the privacy features, it is noticeably worse than Google for searches relating to: locations on a map, programming details, scientific papers and genes/genomics. I find myself navigating to google.com more and more every day. Should I switch back?

#personal experiment #websearch

Is genetically engineered food dangerous? Many people seem to think it is. In the past five years, companies have submitted more than 27,000 products to the Non-GMO Project, which certifies goods that are free of genetically modified organisms. Last year, sales of such products nearly tripled. Whole Foods will soon...

via Jack

Trending Blogs

Recently Viewed Blogs

holycrapscience!