The Snarky Science Cafe

Wherein a mediocre computational biologist chooses "firing squad of science-y bar patrons" to spend his Tuesday night


  1. I convinced some folks at NOVA that I was qualified to give a talk at @CafeSciBoston in front of a decently full room at the Middlesex Lounge in Cambridge.  I wrote a script that I stayed about 65% on, so, to archive my blatherings for posterity, here's what I wanted to say interspersed with the much wiser tweets of the crowd:

  2. Eric Lander is an MIT biology professor and advisor to the president, and he was challenged to summarize the whole human genome project in 6 words. “Bought the book; hard to read.”  Now I’m a post-doctoral researcher at the Whitehead Institute at MIT just up the street and I'm working on understanding this book.  My degree says “bioinformatics,” which means that instead of using chemicals to study biology, I write and use computer programs to process big experimental datasets and come to some conclusion about how your genes work.  

    So tonight, I’m going to tell you about my research on what makes cell types different from each other, how these differences are achieved by molecular machinery, and how we can all do a better job of getting genetics and genomics knowledge to the public.  

    I’d like to practice what I preach by incorporating social media, so to that end, my Twitter handle, @SnarkyScientist, is on my slides if you want to tweet at me during, or bump up my follower count *wink, wink.*  The local branch of Twitter-savvy science communicators, ScienceOnline Beantown, is here, so I expect sparkling conversation under the hashtag, and if you are interested in communicating science on the web, they are the people you should meet.  
  3. My day job is working under Rick Young at the Whitehead, and we look at how cells acquire and keep their identities.  Identity here means what kind of cell something is and what it does, like red blood cells carry oxygen. This is a pretty fundamental question, especially when cancer cells arise from normal cells that have their identities fouled with.

    I’m presuming most of you have had sex ed and/or sex, so I’ll forego the gory details, but you got half of your DNA from your mother, half from your father.  Your cell (singular) divided again and again, and each cell got the same DNA-- the same set of genes or “genome.”  Anything with -ome is just short for “the set of.”

    Now, at a certain point, these cells started to change from stem cells, which are ambivalent about what they want to be when they grow up, into blood cells and skin cells.  How does that happen if they all have the same set of genes?  The increasingly popular but nebulously defined field of epigenomics tries to tackle this question of “many cell types vs. one set of genes.”
  4. Think of DNA as a cookbook.  Each cell has one pristine, mint-condition, Julia Child-autographed cookbook as DNA.  Chunks of this DNA are individual recipes.  Now, you’re not going to use  your precious cookbook for everyday cooking.  No, you make a copy of the recipe.  This working copy of a chunk of DNA is mRNA.  Same instructions, slightly different form.  Like you follow a recipe to make a dish, cells follow the instructions in the mRNA to make a protein.  Proteins form stuff, they make reactions happen, they act as signals.  Some proteins are also are the things that make the copies.  And that is The Central Dogma of Molecular Biology ology ology.  DNA genes are transcribed into mRNA, which is translated into protein.  

  5. Well, not every piece of DNA is a recipe.  There are 3.4 billion letters in 6 feet of DNA in every cell, and about 4% of them have code that make proteins.  Since cells are acting on their own,  DNA also has instructions for when to transcribe genes.  This chunk of DNA is the SOME gene.  It can be transcribed into SOME mRNA and translated into SOME protein.  Other regions of the DNA, called enhancers, get bound by certain proteins and summarize that a gene should be transcribed.  Maybe SOME protein binds this enhancer and causes the SOME gene to be transcribed in a positive feedback loop.
  6. Not all enhancers are created equal.  Our lab recently discovered regions of DNA we call “super-enhancers.”  These are big groups of enhancers that cause genes to be highly transcribed.  The protein-coding genes they control tend to define specific cell type.  Something like “Irish pubs have wooden bars.”-- the fundamentals of what identifies a restaurant type or cell type.  
  7. A Nobel prize went to folks that identified the really important genes in embryonic stem cells. We saw that the transcription of these genes is controlled by super-enhancers.  Then we compared healthy cells to cancer cells and saw new super-enhancers were formed in cancer cells.  The protein-coding genes around these new super-enhancers were really important cancer-associated genes.  So super-enhancers are large clusters of transcriptional enhancers that drive transcription of the key genes that control and define cell identity.  This concept is starting to catch on and we’re seeing papers published by other labs.

    We as scientists have been highly trained in thinking in terms of the scientific paper.  Getting scientific manuscripts published is the sole reasons for our existence.  The number of articles and the prestige of the journals they’re published in is supposed to be a measure of the quality of your science, and we’re rewarded in proportion to this.  So when it comes to communicating our findings outside of the “standard publication” pipeline, we lack some important skills that have come back to bite us.

  8. Let’s go to an authority on the issue: a webcomic.  Ph.D. Comics got me through some dark times in grad school.  And this is their illustration of how a finding goes from benchtop to the public.  Before you think this is hyperbole, it isn’t.  
  9. Here’s what was published in a peer-reviewed journal: “The 5-HTTLPR Polymorphism in the Serotonin Transporter Gene Moderates the Association Between Emotional Behavior and Changes in Marital Satisfaction Over Time.”  UC Berkeley’s press office releases a digest called “Wedded bliss or blues? Scientists link DNA to marital satisfaction.”  So there’s still content there-- it’s an explanation, it’s a digestion of what the paper says and the senior author was quoted.  Then this happened.  I wish I was making this up.  And I wish it wasn’t so commonplace.