Hopefully, this should clear up many misconceptions on the scientific method, along with our favorite post on the definition of a theory (no, it’s never going to “graduate” into a Law!)http://goo.gl/4xqQIf
Fareed Zakaria explains some ideas behind conspiracy theories.
Science is a process guided by simple set of rules scientists follow to make sure that what they do actually works. We have learned over the past four centuries that these are the bare minimum common sense rules to follow. Anything less and you will make mistakes and mislead people into believing falsehoods. Cold hard experience has taught us this over the generations.
Too many people believe science is like some kind of religion, where people just hypothesize and decide that their hypotheses are true, and believe in these hypotheses dogmatically. Evolution and climate change have been specifically targeted by politicians who want people to believe this about science, and all of science suffers as a result of this misinformation. All of science suffers when more and more people misunderstand what it is.
Quote:
1. Make an Observation — “What is happening?”
An Observation is when you notice something in the world around you and decide you want to find out more about it.
2. Define the Question — “Why is this happening?”
Defining the question creates an idea that can be tested using a series of Experiments.
3. Form a Hypothesis — “I think this happens because…”
A Hypothesis is a statement that uses a few Observations, without any experimental evidence, to define why something happens.
4. Perform Experiments — “Let’s test my Hypothesis…”
An Experiment is a series of tests to see if your Hypothesis is correct or incorrect. For each test, record the data you discover.
5. Analyze the Data — “Was my Hypothesis right?”
Analyzing data takes what you found in your Experiments and compares it to your Hypothesis. If needed, perform another Experiment to gather better data.
6. Conclusion — “Experiments show my Hypothesis was…”
Forming a Conclusion presents the Experimental Data and explains how it supports or rejects the Hypothesis. Often, Scientists will take this Conclusion and perform other Experiments on it to discover new things.
(end quote)
7. Request Peer Review — “Did you get the same answer as me?”
Ask other scientists to perform the same Experiments you did to check your work and make sure you didn’t make mistakes, see if they come to the same Conclusion as you did. The more people who get the same answers as you, the more confidence everyone has that you are right.
(thanks to Earl Matthews for sharing this to my stream)
With recent outbreaks of measles, pertussis, and even polio across the United States, I think it’s time to revisit our laws about vaccination.
Refusing to vaccinate your children doesn’t simply put them at risk: it puts at risk everyone who, for medical reasons — extreme youth, old age, health, allergy — can’t be vaccinated, as well as everyone whose vaccinations have lost potency over time. Personal objections to vaccination, whether it be because of (unsubstantiated) fears of side effects or religious reasons, put everyone around you at risk of death or serious injury. This is a classic case where your right to swing your fists around ends at my nose: the individual liberty interest in allowing people to decide which medical procedures to undertake is outweighed by the safety and survival interest of those around you.
I believe that it is time to end all non-medical exemptions from critical vaccination requirements such as MMR, DTaP, and polio — for epidemic diseases which kill and maim by the thousands when our immunity, as a population, is compromised.
My preference would be to treat this as a criminal matter: to fail to do this amounts to reckless endangerment. (Public endangerment, that is, not simply endangerment of a minor) More important, however, is the prevention of harm from people who do this: in particular, individuals unvaccinated without medical reason should be barred from all public accommodations where their presence could put other lives at risk, including schools, parks, pools, and transit.
Such a barring would, of course, have a nearly-catastrophic effect on the life of anyone not living in a remote, rural area; it essentially would reduce a person to second-class citizenship. However, I believe that this is a reasonable accommodation of the public safety interest, as by definition it is something which the person can circumvent by simply not putting the general public in danger by their mere presence.
There are times when I’m willing to be fairly forgiving. When the rightness of a course of action is unclear, I’m generally in favor of letting said action be a matter of individual conscience. However, when an individual’s actions put those around them at risk, this is exactly what we have laws for. You have no more personal right to expose others to deadly diseases than you do to fire a gun blindly into the street.
Like many of the science hype articles in the media, the picture below is misleading. In Buddhini Samarasinghe’s post about cutting through the hype about “exploding” cancer cells: “Exploding Cancer Cells” Explained (http://goo.gl/kZMpVM), we talked about the vehicle control, dimethyl sulfoxide or DMSO. The vehicles below are dangerous but we aren’t talking about that kind of vehicle.
∇ What is a vehicle?
So in biomedical research, what is a vehicle? The American Heritage Medical Dictionary defines vehicle as a substance of no therapeutic value that is used to convey an active medicine for administration. So a drug in a liquid form may use saline (salt water) or a liquid buffer as a vehicle, i.e., the drug is dissolved or mixed with the vehicle. This maybe necessary to get the dose right, i.e., dilute the drug/compound. It maybe necessary to use a vehicle because the compound needs assistance to be transported depending on the route of administration. So for skin creams (transdermal drug delivery) a lotion-like vehicle might be used.
∇ Vehicle Control
No I’m not talking about Automatic Brake Systems or sway-bars. Vehicle controls in biomedical research means a group of test subjects that are given the vehicle alone. It’s like a placebo group but is more specific than that. You’ve probably heard of sugar pills being used for the placebo effect. A vehicle control tests to confirm that the vehicle has no effect on its own. Imagine if you are testing a new drug without a vehicle control group and all of the subjects get sick. You don’t know if it is due to the drug or the vehicle. Similarly, what if all of the subjects show improvement but there is no vehicle group to test to demonstrate that it was the drug alone. In the study that Buddhini Samarasinghe discusses, DMSO was the vehicle.
∇ Dimethyl Sulfoxide (DMSO)
DMSO is often used as a vehicle because a lot of drugs are not water soluble and but are soluble in DMSO. If you want to get a drug into the blood stream, it’s best if it is water soluble. However, some drugs are hydrophobic (they don’t like water but they like oil). If a drug is promising enough, you don’t let hydrophobicity stop you. The LD50 of DMSO is 13.4–15.5 g/kg (12.2–14.1 ml/kg). What does that mean? The LD50 is the dose at which half of the subjects die. LD stands for Lethal Dose. Unfortunately the authors don’t say what dose they used for the vehicle. We don’t know if the vehicle could have had an effect alone. We do know that DMSO can have effects even at low doses.
For example Julien et al found that DMSO had an effect on some enzymes they were interested in for Alzheimer’s disease. They state: These data should caution researchers working with DMSO as it can induce artifactual results both in vivo and in vitro. Galvao et al reported that even low doses of DMSO had toxicity. Finally, Hanslick et al discuss DMSO producing apoptosis in the central nervous system. If you have been paying attention to Buddhini’s Hallmark of Cancer series, you’ll know that apoptosis is programmed cell death.
A couple more comments about the “exploding” cancer cell paper. In one part they show that tumor volume is reduced with treatment and not with DMSO. Tumor volume alone, can be misleading. There are drugs that kill the tumor but the tumor does not shrink, at least not right away. So if the tumor stays the same size, the drug did not necessarily fail. You need functional imaging to show that the tumor is still viable (regardless of size) or the tumor is dying. Also some drugs can make the tumor swell with fluid but the tumor is nevertheless dying. That is another example where tumor volume alone, is misleading. The second comment about the paper is that the live cell imaging experiments were done with an Operetta system. I recommend you check out the video Watch Operetta Product Overview Video (http://goo.gl/FUyKLu) It’s on the right side.
A couple of my favorite quotes are applicable here:
Alle Ding’ sind Gift, und nichts ohn’ Gift; allein die Dosis macht, daß ein Ding kein Gift ist.
“All things are poison, and nothing is without poison; only the dose permits something not to be poisonous.” Paracelsus
The only real difference between medicine and poison is the dose….and intent. Oscar G. Hernandez, MD
∇ References:
LD50 of 13.4–15.5 g/kg (12.2–14.1 ml/kg)
Caujolle F, Caujolle D, H B, Calvet MM (1964) [Toxicity and pharmacological aptitudes of dimethylsulfoxide]. C R Hebd Seances Acad Sci 258: 2224–2226.
Farrant J (1964) Pharmacological actions and toxicity of dimethyl sulphoxide and other compounds which protect smooth muscle during freezing and thawing. J Pharm Pharmacol 16: 472–483.
I dig dragonflies. They have gorgeous colors and they just look awesome. Did you know that their 30,000 lens eyes can also detect ultraviolet light? Because dragonflies and their cousins damselflies don’t have glomeruli, it was thought that they can’t smell. Glomeruli are a cluster of nerve endings near the surface of the olfactory bulb (which is responsible for olfaction, aka smelling) in the brain. I’ve got a cool MRI of a rabbit brain that shows how big the olfactory bulbs are in a rabbit. I should dig that up. Back to the dragonfly, it was recently discovered that they have tiny bulbs in pits on their antennae that may be related to smell. As if dragonflies smelling from their antennae isn’t cool enough, these pits were found using an electron microscope. To test their theory, they used a wind tunnel and dragonfly bait, aka fruit flies. You can read more here:
Dragonflies Lack ‘Smell Center,’ but Can Still Smell
I mentioned that in addition to the minimal amount of blood used, there is a minimal amount of details. Searching elsewhere, even the Theranos website, there aren’t many details. In Di Cleverly’s post, the only decent info was from patents. If you’ve ever read a patent, then you know that it’s often difficult to sort out what’s really going on. So, with Di Cleverly’s help, we have a better picture of what’s going on. A lot of this post are my guesses about some of the details, partly because I’m busy, partly because I’m lazy, and partly because there isn’t a lot out there without really digging. Did I mention I’m lazy, I mean busy?
☼ What was mentioned: small volume and centralized facility
For those that haven’t seen the WIRED or Medscape articles, Elizabeth Holmes dropped out of college at Stanford at the age of 19 and eventually started Theranos with her college funds. The interview talks about how the small volume of blood, from a pin prick, can make the experience, and therefore patient compliance, better. Ms. Holmes talks about reduced and transparent pricing. Essentially none of the technology is discussed. A centralized facility is mentioned. So is that an essential part, i.e., how much can be done off site (e.g. at Walgreens)? Before any young readers decide to drop out like Ms. Holmes or Bill Gates, I think Dave Thomas, founder of Wendy’s makes a good example.
Thomas, realizing that his success as a high school dropout might convince other teenagers to quit school (something he later claimed was a mistake), became a student at Coconut Creek High School. He earned a GED in 1993.)
On Di Cleverly’s post some detective work was done and a few things came to light, mostly via the patents. The small “nanotainer” is used in a novel centrifuge to get information about the blood sample. Red blood cells (RBC) are called erythrocytes and are just one component of blood. If you put whole blood in a glass tube, eventually the RBCs will sink to the bottom and the plasma will stay at the top. You can speed up this process by using a centrifuge (a device that spins the tubes at many times the force of gravity). The rate that the RBCs go to the bottom is called the erythrocyte sedimentation rate or ESR. ESR alone can tell you something about your health.
An increased ESR rate may be due to:
Anemia
Cancers such as lymphoma or multiple myeloma
Kidney disease
Pregnancy
Thyroid disease
Common autoimmune disorders include:
Lupus
Rheumatoid arthritis in adults or children
Very high ESR levels occur with less common autoimmune disorders, including:
Allergic vasculitis
Giant cell arteritis
Hyperfibrinogenemia (increased fibrinogen levels in the blood)
Macroglobulinemia – primary
Necrotizing vasculitis
Polymyalgia rheumatica
An increased ESR rate may be due to some infections, including:
Body-wide (systemic) infection
Bone infections
Infection of the heart or heart valves
Rheumatic fever
Severe skin infections, such as erysipelas
Tuberculosis
Lower-than-normal levels occur with:
Congestive heart failure
Hyperviscosity
Hypofibrinogenemia (decreased fibrinogen levels)
Low plasma protein (due to liver or kidney disease)
The patent mentions a novel centrifuge device with either video or still images of the sample. There are two greyscale figures from the patent in the album below. With image analysis the ESR can be measured without human intervention which minimizes errors.
☼ Microfluidics
Another patent talks about microfluidic devices. I’m assuming those are lab-on-a-chip (LOC) devices. LOCs use microelectromechanical systems (MEMS) to do analysis on very small volumes of fluid. Here’s an example from Harvard that captures trace amounts of tumor cells.
Although genechips or DNA microarray’s aren’t LOCs, it is possible they are being used by Theranos. An image of an Affymetrix Genechip is included in the album below. http://goo.gl/GpMyjx Note the small Eppendorf tubes in the foreground. Those are larger than the Theranos “nanotainer” but they do make Eppendorf tubes the same size as the “nanotainer”. Both the “nanotainer” and Eppendorf tubes have conical bottoms to facilitate removal of all of the liquid. The genechips have target DNA probes attached to the device. If a target gene is expressed, it will bind with the probe on the chip. The readout is typically some type of light whether chemiluminescence, fluorescence, or some combination. The amount of information from these genechips has caused an explosion in bioinformatics and computer processing dedicated to speeding up the analysis of these microarrays.
Because the samples are going to a centralized facility, it’s possible that real-time polymerase chain reaction (RT-PCR) is also being used. RT-PCR is a technique that is used to amplify DNA samples.
☼ Therapeutics and Diagnostics = Theranostics
I didn’t find any information to suggest that the name Theranos has anything to do with the term theranostics, i.e, therapeutics and diagnostics.
Pharmacogenomics aims to identify the genetic basis of variability in drug efficacy and safety, and ultimately develop diagnostics that can individualize pharmacotherapy. Theragnostics, a term denoting the fusion of therapeutics and diagnostics, is receiving increasing attention as pharmacogenomics moves to applications at point of patient care.
Shifting emphasis from pharmacogenomics to theragnostics
An example of theranostics from my boss and colleagues is a platform that combines doxorubicin (cancer therapy), herceptin (targeting for diagnosis), and DOTA-Gd(III) (for MRI detection, i.e, diagnosis). So the herceptin targets the product to cancer cells. Gadolinium, chelated to the construct (DOTA-Gd(III)) allows you to see it with MRI (enhances the contrast from background tissue) and the doxorubicin provides therapy at the target (tumor).
pH-Responsive Theranostic Polymer-Caged Nanobins: Enhanced Cytotoxicity and T1 MRI Contrast by Her2 Targeting
So that’s what I could sort out with the help of Di Cleverly’s post and my own digging through a couple patents. If you have ideas or comments, feel free to ask.
where I mentioned that journal impact factors should be discussed as they play a role in accessing the quality of a journal. Some Open Access journals are good and some are not so much. How can you tell? There’s some nuance and disagreement about impact factors but I’ll get to that later.
First, I want to give a little background and continue the conversation about Open Access journals. In Brent’s post, he mentioned predatory publishers and that we have all gotten spam from them, i.e., requests to consider Open Access journal X when we publish our next manuscript. One of the negative sides of predatory Open Access that I’ve experienced is related to peer review and the role of the editor. After you have done your job reviewing a manuscript and recommended whether or not the manuscript should be accepted for publication, sent back for major revision or rejected outright, the editor takes into consideration the recommendations from all referees and informs the author(s) of his/her decision. The problem is that some predatory Open Access journals charge a significant amount to the authors, sometimes more than $1,500. In the case that I am thinking of, the manuscript was poorly written and was essentially what is known as a quick communication that was being submitted as a full research article. The manuscript was very verbose to try justify full article vs. quick communication. The editor kept pushing to accept the article and to accept it as a full research article. I can only guess the motivation for that was the fee that is charged to the author(s).
Removing the issue of predatory journals, how does one assess the quality of a journal and more importantly a specific journal article. I’ll discuss an example. You probably hear scientists on G+ request peer reviewed citation when “debating” with people. I put debating in quotes because people often don’t know what it really means, it does not mean arguing but I’ll save that for another post. In a debate with a commenter on one of my posts (sorry I couldn’t find the comment to link), he finally gave a link to a peer reviewed article in Bulletins in Insectology. I’m not an entomologist so I have no idea of the accuracy or impact of that particular article. So what do you do?
Phone a friend
Like the Who Wants to be a Millionaire show, one option is to ask an expert. Maybe you know an entomologist. One of the great things about G+ is that you might actually have one in your circles. Alas, I don’t know any or at least couldn’t think of one. The next option is to assess the quality of the journal using impact factors.
Impact Factor (IF)
Journal Citation Reports are made by Institute for Scientific Information (ISI) and can be found on the ISI Web of Knowledge site, owned by Thomson Reuters. You have to have a subscription to the site so this ties in with the Open Access discussion from an access point of view as well. Impact factor is a calculation of the average number of citations per paper for a 2 year period divided by the total number of citable items. The idea of IF is that it gives you an idea of the average importance or impact of articles for a journal. You can imagine where there can be problems with this, e.g., what if a journal publishes a low number of articles per year? The Wiki below goes through more explanations and some alternatives like Page Ranking. A good example in the Wiki is an article that was cited over 6,000 times, yet the other citations for that journal are much lower.
Getting back to the Bulletins in Insectology journal, I looked it up in the Citations Reports. It’s impact factor is 0.44. In the post with the “debate” I had referenced an article in the Proceedings of the National Academy of Science (PNAS). It’s impact factor is 9.737. Just for reference, Science has an IF of 31.027 and Nature has an IF of 38.597. You’ll see them along with other details in the citation report, in the attached figure. Here’s the problem, most people agree that you can’t really compare IF from different disciplines. One reason is that some research might take longer to complete and publish. So if one discipline churns out more publications, that will affect the IF. The number of articles from Bulletins in Insectology is only 42. Remember, IF divides by the number of citable items and a low number should help.
Entomology is a reasonable category to compare Bulletins in Insectology with other journals, in the same discipline. In the next figure below, you’ll see a screenshot of the first page of results (sorted by IF) for journals in entomology. The range of IF is from 13.589 to 1.926. Without being able to “phone a friend”, one would conclude that Bulletins in Insectology is either an obscure journal, new journal, or one that is not ranked high in entomology.
So the Bulletins in Insectology article that was linked is peer reviewed, which is good, but it likely has some issues preventing it from being published in a better journal in the field of entomology.
In our own fields of research, we often don’t pay too much attention to IF because we know which journals our peers/colleagues are publishing in. Unfortunately some academic administrations will use impact factors to judge the quality of someone’s track record for promotion purposes. Again, if they are not in your field, it is one way to assess quality, albeit with the caveats mentioned.
If only the average Joe or Jane could experience having their manuscript ripped to shreds during peer review. Sometimes it’s legitimate and sometimes it’s just a referee that woke up on the wrong side of the bed. Either way, the author has to suck it up and genuflect.
Brian Koberlein explains why science is humbling.
#ScienceSunday
Originally shared by Brian Koberlein
Humility
Yesterday’s post about the big bang and cosmic origins struck a few nerves. Responses ranged from vulgar insults to dismissals of the post as “just a theory.” But more subtle were the criticisms that declared the post lacked humility. Scientific knowledge is never perfect, and to claim the validity of the big bang is to go too far. When communicating to the general public scientists should never say “we know”, only that “we might know.” Scientists should show more humility.
Such criticism fails to recognize that the power of science is its humility. In fact, the scientific process is based on the assumption that individual scientists won’t easily show humility on their own, so it is imposed upon them. There are three basic tenets of scientific research: it must be based upon verifiable data, it must be done publicly, and it must be open to criticism.
Most people view scientific evidence as repeatable experiments that can be done in the lab. For this reason the findings of evolution or cosmology are often countered with “you weren’t there.” But verifiable data is much broader than simply lab experiments. It is a process of gathering data that clearly documents when, where and how the data was gathered. If you gather observational data, the burden is on you to document its origin. If you use data gathered by others, you must clearly cite your sources.
Once you have your observational results or theoretical work, the next step is to present it publicly. This could be a conference, a preprint archive, a book, or submission to a research journal. A scientific discovery is meaningless if it isn’t disseminated. Publication provides a record of the work, so it can’t be tossed down the memory hole. Make a significant discovery, and the record is there. Make a foolish claim, and that’s there too. It’s the latter possibility that strikes fear into scientists everywhere, because publishing your work isn’t sufficient. When you make your research public your colleagues now have a chance to pull the work apart and see if it really says what you think it says. It gets subjected to peer review.
Peer review can be the most frustrating and most humiliating aspect of scientific research. That’s why it’s considered the gold standard of science. Having research published in a peer-reviewed journal means that the work has been examined by other experts in your field, and has been found clear and without obvious error. It doesn’t mean its perfect, but it does mean the work has been held to a high standard and survived. This is why when I write about new scientific work I focus on peer reviewed articles. When I write about work that hasn’t been peer reviewed, I clearly say so.
Of course even after conducting your research, organizing your results, checking it with friendly colleagues, presenting it publicly and submitting it to peer review, you still aren’t done. You’re never done, because at any time someone can critically review your work again. If you have a great theory and your predictions don’t support new findings, we look for something better. No matter how famous, or how many awards you may have, anyone can be toppled by new scientific discovery.
That’s the deal. Keep pushing back against ideas. Keep working to develop better theories. Always, always keep in mind that your theories might just be wrong.
What survives is an understanding of the universe that it robust. It is a confluence of evidence that supports a deep theoretical framework. It is knowledge humbly gathered, and put forward with humility. Through a process that recognizes human fallibility. It is humanity’s best understanding of what is real and true about the cosmos.
This is why I present ideas like the big bang with the claim that we know. We Know. We know because thousands of individuals have devoted their lives to understanding the universe. Devoted their lives to getting it right. Relying on a process that forces us to be humble, and forces us to defend our ideas over and over.
In my posts I always strive to present our best understanding of the universe in a way that is clear and meaningful. That’s why I try to limit moderation of the comments. It is a kind of peer review. I write about science to the best of my ability, and everyone is free to criticize it. I’ve made mistakes in my posts and been called on them. I’ve been praised and thanked for making things clear. I’ve also been called a liar. A fool. Prideful. Deceitful. Ignorant. Arrogant.
An Academic Valentine: Blue for you or Pretty in pink?
Rajini Rao’s #AcademicValentine reminded me of this post about how pH can determine the color of Hydrangeas. Enjoy some science on St. Valentine’s day.
An Academic Valentine: The Science Behind Flower Color
About week ago I posted some pictures of my Hydrangeas that were just starting to bloom. http://goo.gl/Gn47h I noticed that on the same plant, some of the flowers were blue and others were pink. I knew that pH played a role but I found out that it is actually the aluminum in the soil that make the blue pigment possible. So for ScienceSunday curated by Allison Sekuler Rajini Rao Robby Bowles and me, I had to dig up more info to post along with pictures from today.
When the pH is acidic, aluminum in the soil, mostly from clay, allows a metal complex of aluminum and a anthocyanin, named delphinidin 3-monoglucoside, to form. After the pictures, the first figure is of the aluminum complex. The next figure shows various blue flowers with sections cut revealing the pigment cells and protoplasts.
Although the next two figures are about Morning glories, they were too interesting to pass up. A certain ScienceSunday co-curator always has her eyes on certain channels. Similar to the previous figure, there is a cross section-cut revealing the pigmented cells. However, the paper and figure go on to discuss how the Morning glory does not have metal complexation. The petal color changes during flower opening due to pH changes which were measured in the second part of the figure. The final figure show the purported ion channel mechanism.
Plants can be beautiful. When you throw in a dash of science, they can be beautiful and intriguing.
Edit I forgot to add that a lot of insects leave hydrangeas alone. Why? Aluminum toxicity – win – win for us gardeners.
