stenl

Julius Wagner-Jauregg specifically treated neurosyphilis, an infectious disease, not just any ”psychiatric” disorder. That makes it perhaps less surprising that activating the immune system should have an effect. Today neurosyphilis is treated with antibiotics.

How about a nice game of chess?

Neuroscientist here. The study shows no such thing. Maybe you refer to this: ”for the first time, identifies brain tissue macrophages proximal to neurons in over 40% of individuals with schizophrenia who are in a high inflammatory state.”

But first, the authors are confused. All brains (100%) contain macrophages in proximity to neurons. Specifically microglia (which are macrophages) and perivascular macrophages.

Second, note that they studied cases already selected to be in a high inflammatory state.

Third, they do not show the direction of causality. It’s easy to imagine how a life lived with schizophrenia could substantially increase your risk of inflammation.

As a Swede, I have to agree that the level of overt poverty in US cities is shocking (specifically, as it is juxtaposed with so much overt wealth). I don’t know if Anericans care more about helping the poor, but clearly enough people don’t care sufficiently to actually fix the problem by voting for effective social policies. This level of poverty, homelessness and poor health is a political choice, not a law of nature.

I find it very counter-intuitive that they would cite an ”ageing population” as a contributing cause of a drop in life expectancy. I suppose there might be some way that that could be true, but it’s not straightforward.

Solution: free day care. It’s a win for taxpayers if those parents go to work and pay taxes instead of staying home with the kids. Also a win for women’s liberation.

Reminds me of the Ise Jingū shrine, which has been torn down and rebuilt every 20 years for 2000 years.

They should use Bayesian statistics, in which case it doesn’t matter when you stop (more precisely, stopping when you get a result does not bias the outcome; of course running the test longer will make the result more robust). See http://andrewgelman.com/2014/02/13/stopping-rules-bayesian-a...

It’s definitely apparent to those in the field. All the good scientists know who publishes junk, even if the junk publishers are unaware. Arjun Raj has a great and relevant blog post on playing the long game in science: http://rajlaboratory.blogspot.com/2016/06/reproducibility-re.... Worth a read.

One major issue is the fact that reverse transcriptase, which copies the HIV genome inside the cell, is sloppy and makes lots of mistakes. As a consequence, every viral particle produced is a little different. A drug that targets any given HIV protein to disrupt it, will work on most particles, but there’s always some that carry mutations that allow them to escape. The solution is to treat with multiple orthogonal drugs.

However,a second major issue is what’s mentioned in the movie, that the viral genome integrates into the host genome and can lie dormant for years. So even if you kill off every single viral particle, years later new particles can be produced from the dormant genome.

A third major issue is that HIV specifically infects T cells, which are the very immune cells that are supposed to combat infection. This weakens the natural defenses, and also makes it difficult to create a successful vaccine.

I'm not sure what point Bret is trying to make here. Of course interactive explanation can be powerful. As a complement to the actual paper, this can be very valuable, similar to how the HTML version of a paper is useful in many contexts.

But every scientific paper ever published will be around until the end of humanity, whereas this interactive web page will surely disappear within a decade or two. The scientific paper - a self-contained entity that can be moved around, printed out, archived, searched and annotated - is an incredibly powerful entity.

The typical approach is to plot log(x+1) instead, which deals with the zeros as long as you know what you’re looking at. It’s nice to add some jitter too, e.g. a uniform random float number in the -0.25 to 0.25 range.

There is a thing you can buy at the pharmacy that looks like transparent nail polish but tastes extremely bitter. Basically makes it disgusting to bite your nails, but is invisible to others. Hard to get off too, so you’ll get a few days bite-free from one application.

Why is there no internet in the subway in NYC? I always found that annoying.

On the other hand, making kinchi yourself is super-easy. it takes literally five minutes to set up and a few days to ferment, then just gets better over time (weeks/months).

Fun fact: iTunes Store censors actor Brad P++t's last name in Sweden using two asterisks, presumably automatically, presumably because it means d+ck in swedish.

(edit: use + instead of asterisk to fool the markdown rendering)

It's fine to disagree with his political views. That's precisely his point: the freedom to disagree is the foundation of a free society.

What facts were distorted? That he was jailed, multiple times, for defending freedom of expression? That he died in custody? That his wife Liu Xia was under house arrest for years without charges? Or was it his role during the Tiananmen protests and subsequent massacre? That he negotiated the free passage of students from Tiananmen, saving many from the advancing tanks?

I find autopoiesis ("self-constructing, self-maintaining, energy-transducing autocatalytic entities") to be the most satisfying definition of Life. But I find it weird that they would throw that term out there without citing Humberto Maturana or Francisco Varela, who came up with the concept.

They wrote mostly long essays in obscure philosopher's jargon, but one day decided to go ahead and explain everything very simply; they wrote The Tree of Knowledge, which summarizes their philosophy in the style of a children's book. Very worth reading if you're into these kinds of questions. I found a PDF here:

https://www.cybertech.swiss/research/references/Maturana1988...

Luigi from Spotify is very nice. We use it for computational pipelines in our neuroscience research.

https://github.com/spotify/luigi/blob/master/README.rst

Vampire bats use modified capsaicin (chili pepper) receptors in their nose to detect infrared light:

https://www.nature.com/nature/journal/v476/n7358/full/nature...

They use this to find their warm-bodied prey, but of course it's only detection, not vision.

I learned this today, and now I'm happy I did.

Yes. Of course, they should do it properly and not stop at that but look at all plausible factors. For example, there's a very common chromosomal abberration that leads to a ten-fold increase in the violent crime rate (or more), yet this is hardly discussed at all.

So now correlation is causation?

I'm definitely in favour of more high-quality research into the causes of crime and its relationship with immigration, social factors, gender etc. Again, we need facts, which can only be won by serious, honest research, not glib generalizations.

First, to claim that there's a taboo on discussing these issues is ridiculous on its face when it's discussed pretty much every day in every newspaper. You just don't like when people disagree with you that crime is linked to immigration. That link is far from proven. In fact, in the US it's pretty clear that crime does not increase when refugees arrive; if anything it goes down: http://www.citylab.com/crime/2017/02/refugee-resettlement-v-...

I'm not aware of any similar research in Sweden, which is a pity.

I watched Tim Pool's reporting on how media "skews crine statistics" and thought it was pretty good. I would just add that while it's true that some types of crime have increased in the last year or two, we'll have to wait a few more years to know if that's a trend or a natural fluctuation.

Homework assignment for tomorrow: what was the effect on violent crime rates in Sweden of the mass arrival of predominantly muslim refugees from the Balkans in the 1990s?

I'm all for less crime, and for crime stats to be reported accurately. But come on, this is Hacker News, we don't do "well-accepted opinion". Find the facts, report them honestly and then we can have a discussion.

Truth is, crime has fallen everywhere for decades and we don't really know why. Maybe globalization? The banning of lead in gasoline? That's a real mystery worth solving.

Wait, what? The problem is that the facts themselves have a left-wing bias. Crime has fallen over decades, and deadly violence has not changed for a decade (http://www.bra.se/bra/brott-och-statistik/mord-och-drap.html). Real humans experience anecdotes, but it's the job of the media to report the facts. A (large) minority prefers the anecdotes but that doesn't invalidate the facts.

That this breeds mistrust is obviously a problem (maybe THE problem of our era) but the solution can't be to have the media report what your "real humans" want to hear, instead of the facts.

Would some kind soul please divide by the growing size of the population? Meanwhile, this link has population-adjusted charts back to 2006: https://www.bra.se/bra/bra-in-english/home/crime-and-statist...

Off-topic, but Americans literally are shooting each other in the streets. More americans have died in gun homicides than died in all wars since, and including WW2. Include suicides, and gun violence exceeds all US war deaths ever.

http://www.politifact.com/punditfact/statements/2015/aug/27/...

Expensive or not, they didn't do it, so we don't know if those compounds actually work.

Am I missing something or did they also not use a held-out validation dataset to assess performance? It seems to me they ran their autoencoder, got some suggested compounds, listed anecdotal evidence about those compounds, and called it a day.

Those were great! I'm pretty good at doing math in my head and my brain still visualizes numbers and their relationships based on lengths and colors. In 1970s Sweden they were called "tiostavar", "ten-sticks". They were made of high quality painted wood and very tactile and pleasant to work with. The lengths were precise so they would add up properly in any combination, and the widths were equal to one unit of length so you could make areas and volumes.

(Edit) like these: https://orsjoforskoleklass.files.wordpress.com/2014/11/img_3...

Surely the required sample size depends on the effect size. You won't need n>100 to learn that elephants are larger than ants. The problem is in experiment design and interpretation generally, not sample size specifically.

That said, all of us who know how to read a scientific paper (and have institutional access to the journals) should take 5 mins and point out flaws when we see them.