Conference Report: NIH "Human Microbiome Science: Vision for the Future" conference, Bethesda MD

I got back from this conference a couple a little while ago. I had the opportunity to present a poster of some of my work, and a chance to hear from some of the biggest names in the field of microbiome research. You should be able to watch individual talks as soon as they put them up here.

The idea behind this conference was to get feedback from scientists in the area about what the current state of our knowledge is, and to hear about what challenges and difficulties lie in the way of further progress. It was an impressive effort from the NIH people to figure out why they should be funding microbiome research and where the funding would do the most good. With the first phase of the Human Microbiome Project drawing to a close, it was a good chance to stop and organize and plan out what comes next.

I learned a lot about how NIH funding works, what the sub-institutes are, and what they fund. I got to meet a program officer and hear a talk from NIH Director Francis Collins, he of the Human Genome Project fame. I also got to hear talks from such luminaries as Rob Knight, Curtis Huttenhower (Nice guy!), Peter Turnbaugh, Ruth Ley, and Ed Yong (one of the best Twitter follows I ever made). My favorite talk was probably by Maria Gloria Dominguez Bello, who talked about the microbiomes of uncontacted indigenous tribes in Venezuela.

This was also my first time in Washington D.C.. I took the metro downtown one night to have dinner with my PhD advisor and his old PhD advisor (my grand-advisor?), Gary Dunny. Meeting him was a cool opportunity. I liked the atmosphere of the part of D.C. we went to.

I really think there is a bright future ahead in microbiome research, with discoveries just being made about the interactions of the microbiome with hormone regulation, autism development, new probiotics, immune system regulation and autoimmune disease, fecal transplants (So relatively unstudied! So exciting!), and obesity.

And who's to say scientists don't have a sense of humor! My advisor got a chuckle using this sign to talk about correlation vs. causation.

What order these go in makes for very different stories!

Another speaker used this video to talk about the seeking NIH funding in the current funding climate and overcoming challenges along the way. (The cheese is the funding, the grant applicant is the mouse)



I'll be speaking at two conferences between now and October, so this is not the end of conference season for me. It's nice to be able to do conferences at this stage of my PhD and talk about my data.

Book Review: The Signal and the Noise : Why So Many Predictions Fail – but Some Don't. By Nate Silver

How Politics, Sports, and Microbial Ecology are very much alike:

I thought this might be a topical post in light of Nate Silver's announcement that he will be moving his operation to ESPN. Nate is one of my favorite people, as his interests  (Sports, Politics, Big Data) match my own in many ways.

In the field of microbial ecology we are increasingly dealing with mounds upon mounds of data. This is due to the advent of DNA sequencing technologies that can count millions of pieces of DNA and tries to match them to databases that tell us which microbe the DNA came from. Sure, there is signal in these mounds, but there can also be lots of noise. When you make so many observations, there are bound to be some that happen by chance. Even if you are 95% certain that your observations aren't coincidence, it only takes 20 observations before you would expect one to be spurious (19/20= 95%)

 (Relevent XKCD) 

When I started analyzing my own sequencing data, I realized I needed a much better understanding of statistics to be able to grok what my results meant. I had very little formal statistical training, which is a sad reflection on my high school (where all the smart kids should take calculus, I was assured) and undergrad (required a "calculus for business majors" class, but no stats) programs. Ask me when was the last time I formally took a derivative or integral (my undergrad calculus class-- 10 years ago). Ask me when was the last time I used any statistics (yesterday). I think there is a fundamental disconnect between which math skills are actually needed by the majority of people, and which are taught in schools.

Because I didn't have a good foundation on things like Bayes' Theorem, I started looking around for a book that would teach me some fundamentals so I could develop a good feeling for what type of statistical tests would be most appropriate for my data. I didn't want to read a dry textbook. I heard about this book on an interview that Nate did on some TV show and thought it might be an interesting way to learn some statistics. I knew about Nate from his work in predicting elections (the 2012 US elections in particular) and some of his work in sports as well. 

This book talks about the advances in predictions in fields ranging from earthquakes and weather to sports, gambling, and politics. Many of these fields have large data sets to draw from, just like microbial ecology. If you think about it, we have been keeping records in baseball for a very long time. If you wanted to ask how left-handed pitchers do against left-handed batters in the 9th inning of tied games, there is probably a decent sample size to look at. 

As a long time fantasy football and basketball player (one of my hobbies) I have played around with sports statistics for a while to try to make better decisions about who to draft when and what trades to make. (Gotta fill that all-important virtual trophy case!) There is a similar problem in fantasy sports, lots of data, lots of noise. Some people swear that 3rd-year wide-receivers are the most likely to break out, since it take players that long to learn an NFL offense. People said the similar things about quarterbacks for a long time, but then Cam Newton, Andrew Luck, and Robert Griffin III came along and blew away the avoid-rookie-quarterbacks meme. When making sit-start decisions in fantasy basketball, "experts" say that all else being equal, you should always start the player who is playing in a game where the teams are worst at defense, since you get more possessions per game to pile up stats. In actual sports games (not fantasy) there is some debate on whether things like "momentum" are real (is a team/player on a winning streak or a hot scoring streak within a game more likely to perform better than they otherwise would?). I assume one of the reasons ESPN wanted Nate was to help viewers/readers figure which of these "mechanisms" is real and which is noise. The data is there, it just takes a trained person to analyze it. 

Politics also has large datasets going back many years. With this data people try to answer questions such as: Are local elections predictive of national trends? When is the state of the economy a predictor of presidential elections? Will a candidate's race play into the outcome of an election? It takes careful analysis to separate signal from noise. (See the Redskins Rule -- when the Washington Redskins of the NFL win their last home football game prior to the U.S. Presidential Election the incumbent party wins the electoral vote for the White House; when the Redskins lose, the non-incumbent party wins). 

Microbial ecology is similar in that we can get large datasets around which to make hypotheses about the way communities work. We can try to see if they are real by breaking down the numbers and testing our theories about how mechanisms work. Instead of altered run/pass ratios in games with inclement weather, we look at altered bacteroides/firmicutes ratios (different bacterial groups) in obese people. Some correlations end up being real (the proposed mechanism actually influences the outcome) and some end up being the microbial version of the Redskins Rule (no plausible way for the outcome of a football game to affect the outcome of the election). The real mental work comes in proposing likely mechanisms for the correlations you observe and designing further tests to see if those mechanisms hold true. This takes "subject-matter expertise." Instead of proposing that 3rd-year wide receives break out due to learning an offence, we propose that the physiological effects of pH cause shifts in soil communities. 

Anyway, I really enjoyed this book. It keeps a light tone, and was a pretty easy read, even for the statistically uninitiated like me. I recommend it for anyone who may want to work with "big data." I give it 5/5 Petri dishes!

RIP Google Reader

Google Reader was shut down today. I am just one of many who have written about this topic, but I still want to put in my 2 cents.

As someone who likes to think of himself as a "high-information" person, Google Reader absolutely changed the way I use the internet. The ability to aggregate all my favorite blogs and news sources into one page was a game changer; it saved me soooo much time.

I can still remember when I discovered RSS feeds 5 years ago and figured out how to use them. I was
waiting for an experiment to finish and obsessively refreshing some fantasy football advice site that I knew was going to post updated rankings at any moment. I thought to myself "If only there was some way for me to be notified when they updated their site." Clogging up my inbox with subscriptions was a non-starter for me... I wanted a way to keep them separate. I remembered seeing those little orange RSS buttons all over the place and decided to look into what they did. The rest is history. Google Reader quickly became my RSS app of choice due to its simplicity and ease-of-use. I didn't need bells and whistles, I just wanted something efficient. Closing down that last functioning tab of my reader feed was like closing the casket of a loved one (in nature, if not in magnitude, of course).

Google's decision to discontinue Reader still puzzles me a little. The best explanation of what happened that I have found is here. For my purposes, however, there still isn't a better tool out there than RSS feeds keep me up-to-date with my list of sites that keep me informed about the things I am interested in. I don't want to miss posts and I don't want to manually check all of the sites constantly. For now I have found comfort in the arms of Feedly, and it has been... okay. I still miss the Platonic Ideal of simple, efficient, clean layouts that was Google Reader.

Story Behind the Paper*: Bif-TRFLP

One type of post that I plan to regularly put on my blog is the "Story Behind the Paper" series, where I talk about what motivated the research in my papers and tells the paper's backstory. In an aggregation of firsts, this first installment will be talking about my first ever first-author paper, "Use of bifidobacterial specific terminal restriction fragment length polymorphisms to complement next generation sequence profiling of infant gut communities." (Published in the Journal Anaerobe v. 19 (2013) pg. 62-69)

Background

I study the gut communities of infants. There are a couple of nice things about infants that make them an important and useful model. First of all, who doesn't like babies? They are cute! No one wants them to get sick, and studying their gut microbes can help us understand how to prevent and treat gastrointestinal (and possibly other) diseases.

See! Look how cute!
By Weird Beard (Happy) [CC-BY-2.0 (http://creativecommons.org/licenses/by/2.0)], via Wikimedia Commons

Premature infants in particular can have lots of problems, including the very nasty necrotizing enterocolitis. It is thought that a large cause of these types of diseases is an imbalance in the gut community. If there is not enough good (or at least neutral) bacteria in their guts this leaves a spot open for bad bacteria to grow. If we put lots of good bacteria in, the good bugs (metaphorically) eat all the food and sleep in all the beds and there is no room for bad bacteria. This is the concept behind probiotics.

One thing that has held back probiotics as a field is that is is hard to know which effects are caused by which factors. There are a lot of variables, lots of things going on in the gut. Everything you eat introduces new bacteria to the environment and new things for bacteria to eat. Different species of bacteria have "favorite foods" that they like to eat, so what you choose to eat can have a big impact on what type of bacteria can grow in your gut. How can you tell when an effect you are seeing is caused by something different from what you are measuring?

Babies are nice because early on, they mostly eat one thing... breast milk! Many health organizations recommend exclusive breast feeding for at least the first several months of life. This gets rid of one variable that confounds studies on adults. (You usually have to pay adults pretty well to control everything they eat if you want to study them. How much would you charge someone for them to be able to tell you everything you could eat for weeks on end? Tell me in the comments... The economist in me would really like to know!)

If you want to study gut bugs, it usually means you need to collect poop. (There are studies that take pieces of the intestine and other similar methods, but they are much harder to do.) Another nice thing about babies is that most parents are already collecting the their baby's poop by using diapers, and are happy to make a few bucks off of letting us take some.

Now to my paper.

One thing the lab I am in works on a lot is trying to figure out what the gut community of babies consists of. The bacteria my paper studies are the most abundant member of the gut community in most infants, and are considered "good bacteria." They are called "bifidobacteria."

Bifidobacteria as seen in a microscope.

There has been a very important methodological development in biology recently called next-generation sequencing (NGS) that can provide us with data on the microbes in an environment.

The original type of Next-Generation Data
http://memory-alpha.org/en/wiki/Data

It does this by sequencing (finding out the what the order of the A's T's G's and C's is) DNA that is extracted from the microbes in the environment and counting the numbers of DNA snippets that match the sequences of different types of microbes. This has been revolutionary and has enabled us to study things we couldn't before. One limitation of the method (at the moment) is that it can only tell us what is there at a coarse taxonomic level. I'll try to explain this by using an analogy to a more familiar environment... your neighborhood! NGS would tell us how many dogs there were in a neighborhood, but it wouldn't tell us whether they were wolves or dachshunds or chihuahuas or St. Bernards. It could definitely distinguish a cockroach from a human from a shark from a bird though. The problem is that sometimes we really care whether gut bacteria are "wolves" or "chihuahuas." You would probably care about that if you were going to move into a neighborhood, right?

My neighborhood, certified shark free since 1954.
http://www.housing.ucdavis.edu/housing/apartments_orchard_solano.asp

Since what we care about in babies is bifidobacteria, I (together with lots of help from lab members) designed and validated a method that will tell us what species of bifidobacteria are in a baby, since NGS won't do that right now. Since different species of bifidobacteria have different genes that do different things, this will let us test lots of different hypotheses about why bifidobacteria are so common in infant guts, what they do in that environment, and what health effects they have. The method involves taking the DNA from the bacteria in the environment and cutting it with special DNA-cutting enzymes called restriction endonucleases. The enzymes cut the DNA from different species of bifidobacteria in different places, which lets us quickly and cheaply tell them apart from each other by looking at the sizes of the pieces.

Next-generation sequencing methods are constantly improving, and might shortly make doing this unnecessary for lots of purposes, but the lab I am in didn't want to wait for that improvement to answer some important questions. We are using this technique as a part of several other studies now.

If you'll indulge me in some inside baseball of science, I want to note that we ended up submitting this paper twice, to two different journals. The first submission was rejected for reasons that (in my opinion, from my communications with the editor) had less to do with the scientific validity of the paper, and were more about it not being "important" enough for that particular journal to publish. This first journal (which is in the same prestige "tier" as Anaerobe) has the right to do that of course, but it slowed down the publication of the paper. We made some very minor changes (emphasizing how the method is complementary to NGS approaches) and resubmitted to Anaerobe. Anaerobe accepted the paper directly after hearing from the peer-reviewers, who had "no substantive comments" (we didn't need to make any changes). Several of my co-authors on the paper commented that that had never happened to them in their publishing careers, to have a paper accepted with no revisions (and they have a combined 40+ years of publishing under their belts.)

I say this not to out of indignation or a need to be publicly vindicated (since everyone usually thinks their papers are good and should be accepted), but to point out the somewhat capricious nature of scientific publishing. It makes me appreciate the PLoS One model of publishing everything technically sound regardless of perceived importance even more. For those of you who want to read my paper, Anaerobe has a pay wall, so unless you have a subscription (or are part of an organization that does) you may have some trouble. Another point for open access publishing. Here is a PubMed link.


*Note- Hat tip to Jonathan Eisen for the idea of a "Story Behind the Paper" series.

Book Review: Guns, Germs, and Steel

Guns Germs and Steel by Jared Diamond

Okay, so it's been around for a while, but I just recently finished reading it. This book seeks to answer the question "Why did some nations/cultures survive and conquer, while others failed and disappeared." Dr. Diamond narrows the reasons down to three main developments that enable some nations/cultures to beat out others. These developments lend the book its title:

• The development of firearms
• Immunity to diseases, particularly ones found among groups living in higher population densities
• The ability to make steel, useful for weapons and other purposes

But this only gets us one step down. Why did some nations/cultures develop guns and steel and resistance to diseases they could then pass to others who weren't resistant? The author traces the chain of causality down to the basic structure of the Earth, with different parts having different types and numbers of useful plants and animal species depending on the lay of the land. The meat of this book is taking the reader on that journey with heaps of well-explained evidence along the way. As an ecologist working on symbiotic bacterial species, I appreciate his survey of "macrobial" species that influenced human evolution. As someone whose work involves looking at the evolution of human milk across the world, I liked the anthropological information he gave as well. And a book that ties the fate of nations in to microbes... well that's just right up my alley.

This book makes my list because not only is it interesting, it also makes some important points. I know of some fundamentalist religious and political groups that try to justify institutionalized racism by saying that God favors one group of people over others, and cite as evidence the history of specific groups being conquered by others. They say that a nation's/culture's current economic state is evidence of some blessing or punishment given by God for past obedience or disobedience. This book gives readers the scientific ammunition to support a (to me) less odious explanation of how the world ended up the way it is. The idea that one race of people is somehow (genetically?) superior to another can be fought with information about the evolutionary history of the area they live in.

Overall, this is one of my better reads from the last few years. I give it 5/5 Petri dishes!

On the Automation of Data Analysis

I spent the day analyzing data today. I rigged up some Excel spreadsheets to do what I wanted them to do, but it wasn't pretty. I swear by Darwin's ample beard that once I am done with the paper I am trying desperately to get out quickly, and will thus have some time, I will write some Python code to automate the whole process I just did and spit me out my results.

Darwin's Ample Beard. I wonder what he would've though about a wonderful math-doing machine that can keep track of and analyze all the morphometric data about finches you can throw at it.

I know enough about Python (thanks to the Software Carpentry course I took) that I can envision exactly what I want my code to do. (If a cell in this column is blank, fill it in with the contents of the cell above it; Add a new column with labels depending on the value of another column; Delete all rows with a number less than this value in this column; etc.) It is just a matter of sitting down and learning the syntax and functions, which is an investment that will definitely pay off over time, but requires a lump cost up front. (Relevent XKCD) This particular data flow is something I can see the lab I am in spending an hour on every month for at least the next 5 years, which according to the XKCD chart means I can spend up to two days writing code to automate it before it becomes no longer worth it. Challenge accepted! As soon as I finish this paper, I promise...

Why I Decided to Write a Blog

I'm a busy guy. I enjoy my life, and have a good number of hobbies I like doing. Why would I take the time to write a blog, when there are no guarantees anyone would want to read it?

Well,  I'm glad I asked. It started with a paper I read in a scientific journal about social media in the sciences by an acquaintance of mine from a neighboring lab at UC Davis, where I study/work (I never did decide which description is more accurate to describe grad school in the sciences, where you get a stipend for learning about and doing science). Holly Bik wrote this great paper in PLoS Biology (a very respected scientific journal) which included a part about blogging which got me thinking about the benefits of having my own blog. I later attended an event put on by her and others about social media and the sciences. I was persuaded that it would be good for me even if few people ended up reading what I wrote.

One factor that convinced me was that it might help my writing skillz. Another is that potential future employers could get to know who I am by blogstalking me, which would tell them more than a formal resume/CV ever could. I have been in a position to hire/advise about hiring people before, and when deciding between the otherwise qualified candidates I know I would have liked to be able to see who someone really is by reading a blog they write before making a decision.

Another reason is that I think this is a better forum for my thoughts than what I had been using: Facebook. I was getting tired of writing thoughtful Facebook posts that probably just annoyed half my friends who don't care about science/my research/my views and just want to see pictures of my pets or laugh at funny memes I share. (I must admit to feeling similarly about some of my friends' posts who are just into different things than I am.) I would put effort into writing a post that I thought was insightful and would inspire debate and get one measly like. Then I would post a picture of a tree or something and get thirty likes and 10 comments. (Not that there is anything wrong with the tree, or that likes are a good measure of something's worth... but still!) This is probably my fault for expecting too much of the medium and not using it appropriately. I figure writing a blog will be a way to separate my personal and professional lives a little more. Through a blog I can reach people who care about the things I do more easily, and include people besides Facebook friends from high school, who may just want to hear about my major life events.

Seriously, this tree is one of my more popular Facebook posts, beating out my analysis of the Supreme Court decision regarding patenting human genes-- getting 7 times as many combined likes and comments. I guess it is a pretty tree though.

I am on Twitter as well (@zactlewis), but I feel that 140 characters and shared links aren't my cup of tea. I like it for some purposes (its great for keeping up on microbial ecology and finding papers) but I worry I am not communicating very clearly on it.

The last reason is that I feel my blog can be a public good. Other people can benefit from my work in reading and sifting through the amount of information that I do (I am a news/info junkie-- just ask my wife) and maybe have to do less of it when I share things they might find valuable. I know I use other blogs this way. I will probably talk about some of my own research after it is published, and telling "the story behind the paper" can be useful for other people who end up being interested in my research.

So here I am. Welcome to my thoughts. Enjoy your stay.