“Does flossing decrease my risk of heart disease?” No, but is it good for you? Yes.

“Does aluminum cause Alzheimers?” Nope.

“Should I start following the Paleo Diet?” Probably not & paleolithic people probably didn’t either.

As scientists, we’re asked these sorts of questions all the time. Although we’re trained to evaluate scientific ideas, it definitely doesn’t take a PhD to judge the latest craze or newest finding. To do it yourself, follow these 5 steps:

 
1. Separate the sales pitch from the science

Almost everyone is trying to sell something. In articles about science, the sales pitch is usually right in the headline. The science is harder to find.

Start by looking for a quote from a scientist. Read the quote but ignore the spin the author put on it. Don’t forget that scientists can have biases too: be skeptical of scientists who don’t acknowledge the limitations of their research and fail to present alternate explanations. Also, check to see who’s funding the research- they might have an agenda too!

In short, read articles carefully and figure out if the claims they make are based on the facts they present.
 

2. Find the data

Some articles will go beyond quotes and actually describe the studies behind their claims. If they don’t, or if you still want to learn more, use Google Scholar to look for the original source. Search with whatever information you have: the names of the scientists, their institution, or the main topic.

You’ll find some articles that are free and easy to read, but many that aren’t. Even if you can’t access or understand an article, you can still find out if the research was published in a reputable journal. Look for the ‘impact factor’ of the journal where an article was published (search for the journal name + “impact factor”). This number is only given to journals that have been around for at least three years and it reflects how often the papers in that journal have been referenced by other scientists. Look for an impact factor of 5 or higher when deciding if a journal is trustworthy.*
 

3. Evaluate the data

Once you find the data, it’s time to evaluate it. Check to see if the scientists behind the original claim, or another group of scientists, have repeated the experiments (and make sure the results were the same each time!). It’s also better if they analyzed hundreds or thousands of people (or monkeys, or cells, or anything else) instead of just two or three. Also keep an eye out for all the differences between two groups in a study, especially with humans. Things like income levels and access to health care can sometimes explain the reported results better than whatever the article is proposing.

Think about it this way: if you were in charge of figuring out the height of the average American male, you would need to measure a bunch of people to get it right. If you only measured a few people, and they happened to be basketball players, you’d be way off.

Also keep an eye out for misleading graphs. Graphs are great for communicating complicated information quickly, but they can also be misleading. Here are a few classic graphical tricks to watch out for:
 

The “put two things with the same trend on one graph” trick:

(source: Businessweek)

This one’s pretty obvious: correlation does not equal causation. It’s exaggerated, but it’s a good example of how putting two things on the same graph doesn’t mean they’re related… even if it looks like they might be.

The “take out data points” trick:

original (left) vs. doctored (right)

Here, we removed the last (very important) column that put the other columns into context. This is a great example of why you should find the original graph from the primary research article whenever you can.

The “zoom in on the axis” trick:

These two graphs have the same information. In the graph on the right, we’re zooming in on the top bit of information, making it seem like the differences between the columns are bigger than they are. Always check the numbers on the left side (y axis), and what those numbers stand for.

Finally, cover up the title, look at the graph and ask yourself what title you would give the data. If you don’t think you can conclude anything from the graph, you should be skeptical of anyone who claims they can. This applies to any kind of information, not just graphs!

 
4. Put the story into context

Now that you know what the study is saying, find out how it fits into the big picture. Make sure there’s more than one paper written about the subject and look for quotes from scientists who weren’t directly involved in the research. Different labs and researchers will often disagree about a scientific topic (which is why you should distrust the phrase “scientists say”). If you’re having trouble finding alternative perspectives, the Wikipedia page for the topic can be a good place to start, especially if it contains a “controversy” or “criticism” section.

 
5. Ask an expert

If after all this research and data analysis you still feel lost, find someone you trust and ask them about it! Is there a science blogger you like? Tweet at them. If you don’t have a favorite yet, ask us. We’d love to help and point you in the right direction toward solid resources. Nothing beats a real discussion (even over Twitter or email!), but you can also check out neutral, non-biased sites like Mayo Clinic.

 
More helpful resources:

• Google has some great tutorials on how to search. They give advice on which terms to choose and how to evaluate the search results.
• The PLoS journal series, especially PLoS Biology, is an open source, peer-reviewed journal series. All the articles are free, and many are written clearly, so read away!
• PubMed is another search engine for science journal articles.
• This guide shows you how to convert obscure technical measurements into simple terms.
• This list of “predatory” journals is a good guide to what not to trust. These journals will publish any research for a hefty price. They have no quality control.
• This list of 10 Questions you should ask to distinguish real from fake science is a great way to decide whether or not to trust a source.
• The UK National Health Service runs an excellent fact-checking website that can help you look behind the headlines.
• Here’s a great article detailing some of the biases and pitfalls in data analysis, especially for small studies.

 
Useful links from the comments (thanks guys!):

• For even more open access journals check out the Directory of Open Access Journals (DOAJ).
• For physics papers, you might have better luck over at arXiv, which archives reprints.  Make sure the article has been “accepted for publication” in a journal, though!
• There’s also a preprint archive for biology papers called PeerJ Preprints.
• Pubpeer.com is a forum where peer-reviewed scientists comment on and critique recent published findings.
• This thorough post helps you tell “a good website from a crap website”.

 
*This threshold is most relevant to biomedical and basic biology journals. Other fields may have fewer citations, but that doesn’t mean that the research is necessarily untrustworthy. Impact factors certainly don’t tell the whole story about a journal’s value. If you want to know more, check out this article that delves into a little more detail.

Happy Pi Day!

 It’s a fact

A ratio immutable

Of circle round and width

Produces geometry’s deepest conundrum

For as the numerals stay random

No repeat lets out its presence

Yet it forever stretches forth

Nothing to eternity.

- David Saul and Danielle Mathieson

 Can you figure out the pattern in this piem (pi + poem – us math geeks love wordplay)? Remember, pi = 3.1415926535897932384626433832795028… It is not a coincidence that the first word word of the poem has 3 letters, the second word has 1 letter, the third has 4 letters, etc.  People use piems as mnemonics (piphilology) to memorize the digits of pi. However, if you want to memorize large sequences of pi, I would recommend more creative methods.

Currently, Lu Chao holds the world record for number of digits recited. Over the course of 24 hours and 4 minutes he recited 67,890 digits of pi. There were no bathroom breaks because the Guinness Book of World Records mandates that you only have 15 seconds between numbers. The youngest pi competitor, so far, is a three-year-old girl, Grace Hare, who recited 31 digits! What a feat. I’m excited when I correctly recite my nine digit social security number.

Pi is the ratio of a circle’s circumference to its diameter. As school children, pi is the first irrational number we learn, and as adults it is typically the only irrational number we remember. The number hold such a prominent place in our hearts because circles dominate both natural and man made worlds.

Circles were pervasive in the architecture, art and tools of early civilizations. As early as 2000 BC, the Babylonians and Egyptians recognized that the circumference and diameter have a constant ratio. A mixture of fascination and necessity drove them to approximate that ratio.

Babylonians: pi = 3 1/8 ≈ 3.125 (only 0.53 % off!)

Egyptians: pi = 4*(8/9)2 ≈ 3.1605 (only 0.60 % off!)

These were pretty incredible approximations considering that at this time, wooly mammoths were still alive, and materials like glass, iron, and bronze were still unknown to humans.  Interestingly, the concept of pi and the concept of zero was conceived of around the same time, though zero took much longer to catch on.

Now, thanks to the work of some dedicated mathematicians, pi has been calculated up to 10 trillion digits. Think of how many years it would take Lu Chao to recite that.

So today, March 14th, let’s all raise our forks and dig into this delicious number, preferably at 1:59! (3.14159 = 3/14 at 1:59)

Monkey poop: a scientific goldmine

Have you ever wondered what mysteries primate poop could unlock? No? Me neither. But luckily, Dr. Beatrice Hahn did. Dr. Hahn was deeply curious about the origin and evolution of HIV (human immunodeficiency virus) so, naturally, her path lead to poop.

Don’t see the connection? Well, during Dr. Hahn’s recent visit to The Rockefeller University, she walked us down the poop path toward some astounding conclusions.

Why monkeys?

When HIV was identified in 1983, researchers almost immediately suspected that the virus came from non-human primates. This suspicion was heightened when a team of researchers found AIDS (acquired immunodeficiency syndrome), the fatal final stage of HIV, in a captive colony of macaque monkeys.

Over the last two decades, Dr. Hahn’s team has shown that HIV originated in a sub-species of chimpanzee. But how did they solve this mystery?

HIV in Humans

HIV is diagnosed in humans by looking for antibodies against the virus in the blood and saliva. When infectious agents like viruses or bacteria enter our body, our immune system generates antibodies that help destroy the invaders. Importantly, each of the thousands of antibodies we make generally recognizes one type of foreign substance. For instance, I got chicken pox when I was three, so if you searched in my blood, you would find antibodies against this virus. Conversely, I thought I had malaria on my 21st birthday, but a blood test revealed that my body was not producing antibodies against it- I was just really sick for some other reason.

SIV in monkeys

Simian immunodeficiency virus (SIV is the monkey equivalent of HIV) is also diagnosed by finding antibodies against the virus. The natural place to look for these antibodies would be the blood. But, fortunately or unfortunately, there are very strict regulations on working with wild monkey populations and Dr. Hahn’s collaborators did not get permission to collect blood. So how did they collect antibodies?

Ingeniously, Dr. Hahn and her colleagues devised a way to find SIV antibodies in feces! Over the past decade, teams of animal trackers across Africa have mailed Dr. Hahn’s lab fecal samples taken from primate populations.

Determining evolutionary relationships

Once Dr. Hahn’s team found SIV antibodies in the poop of many species of monkey, how did she determine which SIV was the ancestor of HIV?

Scientists can determine evolutionary relationships by evaluating the similarity of genetic sequences between species. For instance, when trying to determine our closest evolutionary relative, scientists compared human genetic sequences (our DNA) to ape and monkey sequences. Chimpanzees were found to be humans’ closest evolutionary relatives because their genetic sequences were more similar to ours than to other monkeys. This also means that chimpanzees and humans share a common ancestor.

The evolutionary history of apes. Each branching point represents a common ancestor.

Dr. Hahn used the same logic to determine the evolutionary relationship of the viruses, except instead of directly comparing genetic sequence, she compared the sequence of antibodies made by the host against the virus.

Jumping Barriers

SIV has jumped the species barrier to infect humans in only a handful of cases, but Dr. Hahn’s group found that all HIV came from central African chimps. HIV’s cross-species jump has occurred independently at least four times. One of those jumps was the source of the pandemic M group HIV-1, which is responsible for 90% of the 33 million cases of HIV/AIDS.

While studying these cross-species jumps, Dr. Hahn found it curious that although there were two closely related types of SIV in chimps (one that infects chimps in central Africa and one that infects chimps in eastern Africa), the HIV originating from the central African SIV has been much more devastating to humans.

Why hasn’t SIV from eastern African chimps infected humans?

Dr. Hahn laid out three possibilities.

1.  People were not exposed to SIV from these chimps

Very unlikely. Researchers think that humans became infected with SIV (which then evolved into HIV) by hunting and eating chimps. Both central and eastern chimps are killed for their meat.

2.  Infection has occurred, but just hasn’t been detected

Unlikely, but still possible.

3.  This type of SIV needs to undergo more mutations in order to survive in a human

Most likely. In order to survive and replicate (and cause all the nasty HIV/AIDS symptoms), the virus needs to interact with many host proteins. One barrier to SIV jumping species is that the primate version of a protein can look very different from the human version. If the primate virus cannot recognize the human protein, it cannot interact with it. Therefore, in order for primate virus to start infecting humans, it has to mutate (change its genetic code) to recognize human proteins.

In support of the third possibility, Dr. Hahn’s group found one striking example of a viral protein that needs to mutate to infect human hosts. Interestingly, this viral protein needs to undergo fewer mutations in the central African SIV than the eastern African SIV to become infectious to humans. They hypothesize that the need for fewer mutations made it easier for the central African virus to jump species and become infectious compared to the east African virus.

 Why is this protein special?

Background: HIV and SIV replicate inside a specific type of immune cell (CD4+). To infect more cells, the virus needs to leave the cell it is in. Imagine an infected CD4+ cell as a ship that is being plundered by pirates (HIV). The pirates, after plundering the ship, want to leave and go plunder other ships. Before leaving this ship, they need to cut the ropes that are tethering them.*

HIV and SIV: “pirate-like” viruses that use different protein “knives” to cut their way out of cells

Tetherin, a membrane protein discovered by Rockefeller’s Paul Bieniasz, binds to the HIV virus as it passes from the inside of the cell to the outside, preventing it from leaving. If the tetherin rope is not cut, the HIV pirates cannot leave to go plunder other ships. In order to cut these ropes, they need a very special knife. There are two types of proteins that are effective knives: vpu and nef.

The rope vs the knife: To cut ropes from a chimp ship (CD4+ cell), the chimp pirate uses a nef knife. However, chimp nef knives cannot cut human tetherin ropes. Biologically speaking, this is because the chunk of tetherin protein that nef recognizes has been deleted in humans. The virus that wants to jump ship (from chimps to humans)  needs a new knife to infect humans. The protein knife can only change as a result of random mutation (genetic changes that happen to the virus by accident as it replicates).

The outcome: Unfortunately for millions of people, one of those random mutations (in another protein, called vpu) created a new protein knife for the virus. In humans, HIV pirates use vpu to cut the tetherin ropes and go plunder other CD4+ ships. Each time SIV jumped species to become HIV, a slightly different mutation turned vpu into a good tetherin knife. The vpu of the pandemic HIV is the most effective cutter of them all. Dr. Hahn speculates that this is why this specific type is so infectious.

Interestingly, vpu in SIV from eastern chimps needs to mutate more times to adapt to human tetherin. In other words, the knife needs to change more in order to cut the tetherin ropes. This could be one reason why only central chimp SIV has successfully jumped the species barrier.

Summary

By removing SIV antibodies from African monkey poop and comparing their sequence to HIV antibodies from humans, Dr. Hahn’s group showed that the pandemic HIV arose from chimpanzees from central Africa. Dr. Hahn has gone on to discover many other interesting facts about humans and monkeys. For instance, using similar techniques, she has found that western gorillas are the reservoir of Plasmodium falciparum, the most deadly form of malaria. But that is a story for a different post.

*Disclaimer: I use the word “cut” for illustrative purposes. It is unknown how vpu and nef disrupt the function of tetherin.

This has been brought to you by: