darb.ketyov.com

Caveat lector: This blog is where I try out new ideas. I will often be wrong, but that's the point.

Home | Personal | Entertainment | Professional | Publications | Blog

Search Archive

Loading...

28.4.11

TEDbrain

TEDbrain: Dragging neuroscience into the 21st century

(Source: Neil Girling)


That's me after my TEDxBerkeley last year. If I could do my talk again, it would be very different.

If it were tomorrow, I'd talk about how much more we could be doing in neuroscience if only some of the old walls were brought down. If only there was a way to bring together engineers, visual artists, designers, scientists, philosophers, mathematicians, etc. under one roof.

Not only do I want to see a TEDbrain, I want to see an entirely new research environment.

The mysteries of the brain aren't going to be solved with one, beautiful equation, or by a super-elegant experiment, or by a great new technique. The problem is bigger than that.

Understanding the brain isn't just a scientific problem, it's a data problem.

Granted there have been some amazing advances, but most cognitive neuroscience experiments are basically 1960s psychology experiments with better toys.

We can do better. We need to think bigger. We need to leverage the massive amounts of data we're collecting.

We need better data ontolgies. Better data mining. Better data visualization.

Right now, neuroscience in particular is undergoing a renaissance.

Hell, when you've got people like Sergey Brin who want to help spur on novel research, it strikes me as silly that a lot of the ways we're thinking about brain problems hasn't changed too drastically.

Neuroscience is fairly unique among the sciences in that it pulls in from so many different disciplines.

At Google last year, I gave a presentation titled, Computational Analysis Methods and Issues in Human Cognitive Neuroscience that was supposed to be about the difficulty we have in making the most out of the huge amounts of brain data that we collect from each of our subjects. We run very narrow experiments and throw out most of the information in our datasets by intentionally removing information via averaging, filtering, smoothing, a priori selection, etc.

Wasting data is stupid. But we all do it.

This winter, my wife and I put together brainSCANr in an attempt to synthesize and mine data from published research papers.

Tal Yarkoni and Russ Poldrack have done a more sophisticated job with PubBrain, which is a super cool, massive meta-analytic tool. Tal and Russ are doing an amazing job, but they're a rare breed.

My soon-to-be post-doc advisor, Adam Gazzaley, is doing awesome work in modernizing cognitive neurosciecne. At TEDxSanJose this year, he spoke about his work on taking neuroscience experiments out of the lab and into peoples' homes. About giving them at-home EEG and testing them using more natural stimuli (i.e., more "ethologically valid").

He and I have a really amazing project in the works.

The Bay Area is uniquely positioned at the interface of academic science and the intellectual tech communities. It would be the perfect place for a TEDbrain event.

The academic system around us is falling apart. Grant writing and politics make for a difficult environment for a researcher.

How is a scientist supposed to innovate when they're worried about their next grant? But I digress...

18.4.11

The 10 most cited neuroscience papers

ResearchBlogging.org
Over on Quora someone asked, "What are the most-cited works in neuroscience?"

As some of you may know, I love literature mining. I started brainSCANr with my wife to look at how neuroscience research is conducted. I've been talking with Neuroskeptic about looking at biases in how different neuroscience topics are studied.

This question on Quora piqued my interest, so I decided to try and work out the answer. I had to do some digging.

(source)


My first thought was of a retrospective published in Nature Reviews Neuroscience last year that revisited ten of its most-cited papers between 2000 and 2010: Ten years of Nature Reviews Neuroscience: insights from the highly cited.

A quick Google search showed that the Journal of Neuroscience also has a list of its most-cited papers on its site.

But in both of these cases we've got a clear bias: they only show the highly-cited articles published within their respective journals. And there are a lot of journals that publish neuroscience research.

Any answer I can find will come with a lot of caveats. It's difficult enough to define which papers are "neuroscience", let alone track down the "top cited" ones in an unbiased manner. From what I've seen, the most-cited papers tend to be methodological or crystallography structures. However, in order to reduce any sub-field bias, I wanted to find a database of papers to work from.

The ISI Web of Knowledge runs a site, ISI Highly Cited, wherein you can search by specific neuroscience researcher, but that information doesn't tell you about specific highly-cited papers. Cool, but no good here.

Thomson Reuters (who owns the Institute for Scientific Information (ISI)) also runs ProCite, which, thankfully, has a nice little site wherein they provide citation counts for 2610 neuroscience papers published in the ten-year period between 1995 and 2005. Those data come in a nicely-formatted text file.

Yay datas!

These types of files are data pornography for me. They make what I do so easy and fun!

(Yeah, I said it. Fun. What?)

(source)


Mind you, the results from ProCite come with caveats: I can't say that they're comprehensive (I don't know how they defined "neuroscience") and they're temporally restricted (1995-2005). Never the less, it's a good start.

Of those 2610 "highly cited" articles, 1877 had more than 100 citations. 1951 of the 2610 articles were classified as "research articles", 656 as "reviews" (and 3 as "notes"?) There were a total of 108 unique journals represented in this sample.

The breakdown by of the 20 journals with the largest number of highly-cited articles looks like this:

Bradley Voytek Neuron PNAS citations

(Why yes, I do have a first-authored paper in three of those top five journals! I hadn't noticed! What? You'd like to give me a job? Why thank you! That would be delightful. ::COUGH::)

As an aside, I'd like to point out that 4 of the top 10 journals have one-word journal titles: Neuron, Nature, Science, and Neurology. Out of the 108 total journals represented, 22 have one-word titles. This means we should only see two--not four--one-word title journals in the top 10. This supports my pet theory that one-word title journals fare better than journals with longer titles. (This is a little ironic, because the journal with the longest title is also in the top 10: Proceedings of the National Academy of Sciences of the United States of America, but they've long since learned to refer to themselves simply as "PNAS".)

Of course, the earlier the year of the paper's publication, the more frequently it's been cited, which suggests that the citation counts for these articles will continue to increase (that is, after 10 years they still haven't really stopped getting cited).

Bradley Voytek citations

Without further ado, here are the 10 most highly-cited neuroscience articles published between 1995 and 2005, ordered by number of citations:

1. (1684 citations) Sherrington et al., Cloning of a gene bearing missense mutations in early-onset familial Alzheimer's disease. Nature 1995.
2. (1603 citations) Polymeropoulos et al., Mutation in the alpha-synuclein gene identified in families with Parkinson's disease. Science 1997.
3. (1513 citations) Caterina et al., The capsaicin receptor: a heat-activated ion channel in the pain pathway. Nature 1997.
4. (1396 citations) Südhof, The synaptic vesicle cycle: a cascade of protein-protein interactions. Nature 1995.
5. (1369 citations) Pin & Duvoisin, The metabotropic glutamate receptors: structure and functions. Neuropharmacology 1995.
6. (1288 citations) Tessier-Lavigne & Goodman, The molecular biology of axon guidance. Science 1996.
7. (1246 citations) McKeith et al., Consensus guidelines for the clinical and pathologic diagnosis of dementia with Lewy bodies (DLB): report of the consortium on DLB international workshop. Neurology 1996.
8. (1186 citations) Games et al., Alzheimer-type neuropathology in transgenic mice overexpressing V717F beta-amyloid precursor protein. Nature 1995.
9. (1065 citations) Hsiao et al., Correlative memory deficits, Abeta elevation, and amyloid plaques in transgenic mice. Science 1996.
10. (1050 citations) Levy-Lahad et al., Candidate gene for the chromosome 1 familial Alzheimer's disease locus. Science 1995.

What's amazing to me is that 7 of the top ten papers are from Nature or Science, and 5 of them are more or less about Alzheimer's disease. This suggests a relative homogeneity of "top" researched topics.

As a final note, it's important to point out that these publications are from 1995-2005. They ignore older papers, as well as more recent ones. Given that the modern scientific publishing era began in 2003 with the founding of PloS and PLoS Biology, there wasn't really any time for papers published in those journals to really have any representation in this dataset.

Luo, L., Rodriguez, E., Jerbi, K., Lachaux, J., Martinerie, J., Corbetta, M., Shulman, G., Piomelli, D., Turrigiano, G., Nelson, S., Joëls, M., de Kloet, E., Holsboer, F., Amodio, D., Frith, C., Block, M., Zecca, L., Hong, J., Dantzer, R., Kelley, K., & (Bud) Craig, A. (2010). Ten years of Nature Reviews Neuroscience: insights from the highly cited Nature Reviews Neuroscience, 11 (10), 718-726 DOI: 10.1038/nrn2912
Sherrington R, Rogaev EI, Liang Y, Rogaeva EA, Levesque G, Ikeda M, Chi H, Lin C, Li G, Holman K, Tsuda T, Mar L, Foncin JF, Bruni AC, Montesi MP, Sorbi S, Rainero I, Pinessi L, Nee L, Chumakov I, Pollen D, Brookes A, Sanseau P, Polinsky RJ, Wasco W, Da Silva HA, Haines JL, Perkicak-Vance MA, Tanzi RE, Roses AD, Fraser PE, Rommens JM, & St George-Hyslop PH (1995). Cloning of a gene bearing missense mutations in early-onset familial Alzheimer's disease. Nature, 375 (6534), 754-60 PMID: 7596406
Polymeropoulos MH, Lavedan C, Leroy E, Ide SE, Dehejia A, Dutra A, Pike B, Root H, Rubenstein J, Boyer R, Stenroos ES, Chandrasekharappa S, Athanassiadou A, Papapetropoulos T, Johnson WG, Lazzarini AM, Duvoisin RC, Di Iorio G, Golbe LI, & Nussbaum RL (1997). Mutation in the alpha-synuclein gene identified in families with Parkinson's disease. Science (New York, N.Y.), 276 (5321), 2045-7 PMID: 9197268
Caterina MJ, Schumacher MA, Tominaga M, Rosen TA, Levine JD, & Julius D (1997). The capsaicin receptor: a heat-activated ion channel in the pain pathway. Nature, 389 (6653), 816-24 PMID: 9349813
Südhof TC (1995). The synaptic vesicle cycle: a cascade of protein-protein interactions. Nature, 375 (6533), 645-53 PMID: 7791897
Pin JP, & Duvoisin R (1995). The metabotropic glutamate receptors: structure and functions. Neuropharmacology, 34 (1), 1-26 PMID: 7623957
Tessier-Lavigne M, & Goodman CS (1996). The molecular biology of axon guidance. Science (New York, N.Y.), 274 (5290), 1123-33 PMID: 8895455
McKeith IG, Galasko D, Kosaka K, Perry EK, Dickson DW, Hansen LA, Salmon DP, Lowe J, Mirra SS, Byrne EJ, Lennox G, Quinn NP, Edwardson JA, Ince PG, Bergeron C, Burns A, Miller BL, Lovestone S, Collerton D, Jansen EN, Ballard C, de Vos RA, Wilcock GK, Jellinger KA, & Perry RH (1996). Consensus guidelines for the clinical and pathologic diagnosis of dementia with Lewy bodies (DLB): report of the consortium on DLB international workshop. Neurology, 47 (5), 1113-24 PMID: 8909416
Games D, Adams D, Alessandrini R, Barbour R, Berthelette P, Blackwell C, Carr T, Clemens J, Donaldson T, & Gillespie F (1995). Alzheimer-type neuropathology in transgenic mice overexpressing V717F beta-amyloid precursor protein. Nature, 373 (6514), 523-7 PMID: 7845465
Hsiao K, Chapman P, Nilsen S, Eckman C, Harigaya Y, Younkin S, Yang F, & Cole G (1996). Correlative memory deficits, Abeta elevation, and amyloid plaques in transgenic mice. Science (New York, N.Y.), 274 (5284), 99-102 PMID: 8810256
Levy-Lahad E, Wasco W, Poorkaj P, Romano DM, Oshima J, Pettingell WH, Yu CE, Jondro PD, Schmidt SD, & Wang K (1995). Candidate gene for the chromosome 1 familial Alzheimer's disease locus. Science (New York, N.Y.), 269 (5226), 973-7 PMID: 7638622

4.4.11

How to be an anti-scientist

Along the lines of my "how to be a neuroscientist" post, I bring you: How to be an anti-scientist!

(Source - The Onion)

Example 1: Heliocentrism
Since [the astronomer] cannot in any way attain to the true causes, he will adopt whatever suppositions enable the motions to be computed correctly from the principles of geometry for the future as well as for the past... these hypotheses need not be true nor even probable.

Example 2: Evolution
This textbook contains material on evolution. Evolution is a theory, not a fact, regarding the origin of living things. This material should be approached with an open mind, studied carefully and critically considered.

The top quote is a preface to Copernicus's On the Revolutions secretly prepended without Copernicus's knowledge by a Christian theologian in the 1540s.

The bottom quote is a preface to biology textbooks distributed in Atlanta public high schools in 2002.

That gives us a Δt of approximately 460 years across which the complexity of the anti-science argument has gone from:

"...[T]hese hypotheses need not be true nor even probable."
to:
"[This scientific idea] is a theory, not a fact."

So, how does one succeed as an anti-scientist? Just keep repeating the same thing over and over again for centuries while the world around you changes ever faster.