“Thinking” about change & insulin

I didn’t write a post last night because I got back to my hotel a little late and went to bed. I attended the Division of Comparative Endocrinology social from 8-10 last night. SICB is divided into divisions like this, there’s a whole bunch of them. By breaking up membership by subject of study, it’s easier to organize facets of the meeting and connect people who are interested in the same things. This social was especially fun because they had five masters students do a “data blitz,” a mini presentation in two minutes. One girl studies starling stress and presented her work in verse, the guy that followed her crammed 45 slides into two minutes and if it was not the best presentation I’ve ever seen it was certainly the funniest. He had a whole crowd of tired, drink-in-hand scientists laughing and paying attention to him.

Afterwards, I went out to a close-by bar called “Wanna-B’s” with Eli and several of the people who had stuck around to the end of the social. I saw several super intelligent and well-respected researchers sing, including one duet of “Let It Go” from Frozen. I had the biggest smile on my face. It’s the best thing to know Eli here because he comes to SICB every year. I’m getting to meet and be friendly with people I would never have gotten the chance to without his introduction. It’s clear that SICB is everyone’s favorite meeting by far.

Yesterday’s end was eventful but the whole day was too. For most of the morning, I attended a symposium called “Thinking about change: an integrative approach for examining cognition in a changing world.” This is more closely related to the kind of thing I’m interested in studying in graduate school. There were several talks dealing with stress in the early environment and how experience affects neurogenesis, especially in the hippocampus. I learned about the reactive scope of stress. There is an optimal amount of stress under which the brain responds well, but on either side of this point neurogenesis declines. I learned about something called psychosocial stressors, or perceived challenges that may or may not be as challenging or stressful in reality.

Some thoughts I wrote down during this session:

So, the problem is when a stressor enters the overload range and the HPA axis/mechanisms of neurogenesis can’t deal. If you told someone post-stressful event, would it push that stress back into the predictable/reactive region? Can you retroactively “fix” an extreme stress? What are the ways in which you can manipulate neurogenesis? Can you retroactively change predictability?

I also attended some more talks relevant to my thesis. One was on IGF-1 in the house mouse. The presenter found that IGF-1 did not correlate with reproductive effort, as measured by how many offspring a female mouse gave birth to. This was her prediction, so the take-home message of the talk was that this is just another thing we know that IGF-1 doesn’t do. It seems like this is the path we are on as we figure out complicated and pleiotropic players in physiology. Experiment after experiment that yield, well…. it doesn’t do this. 

That talk used what the presenter called a “linear mixed model with pairwise comparisons.” That sounds an awful lot like what I’ve been using. It was helpful for me to see how she phrased that.

I was really excited to attend an afternoon session on hormones and signaling pathways in development. One of the speakers is in the lab of a paper that I actually cited on my poster. That talk about about the insulin system and it was really helpful for me to hear a summary, because I still feel like I’m struggling to get specific about what “insulin” signaling means.

“Insulin signaling is an organism’s system for receiving information about the availability of food and the quality of that food, and for allocating the energy derived from it to various body parts and systems in order to control growth and development.”

Their results indicated another piece of interesting information – that the insulin receptor may not regulate nutrition-responsive growth, at least in horned beetles.

A guy gave a talk on soapberry beetles that was super cool. Both he and two members of the insulin lab stopped by my poster later in the afternoon. It was really helpful to talk to both of them, which I’ll address in just a second.

The last talk in the afternoon session gave me a helpful phrase for thinking about sexual dimorphism. “The regulation of sexual dimorphism can be coupled and uncoupled from nutrition because sexual dimorphism in clades is not uniform” Meaning that in some animals, sexual dimorphism may interact more interestingly/dependently than in others.

Finally, the poster session. Turns out, I’m only slightly happy with how my poster came out. I like the layout and the sizing, but I wanted the background color to be tan and the headers to be green when really it all came out kinda just… green. It looks okay up close but is not attractive from far away, which is something you want in a poster. Oh well.

Poster thoughts for next time – I like the three panel lay-out. Instead of boxes around my pieces of information, I think I’ll try putting section headings in a block of color with white text. It’s okay for the results section to take up more than one column this way. I think I’ll also stick with white and some other complementary colors. The graphs should be bigger next time and I should get more pictures.

I’m pleased with the way I explained my research. I don’t think the “spiel” was too long, and I rarely said “um.” I felt like it had been a little while since I presented anything, so this was nice.

In the ongoing saga of how to deal with gene expression data and what does that even mean, I had a great conversation with the soapberry bug researcher from Colby College. His name is Dave Angelini and I asked him about how he thinks expression data should be normalized. He said he thinks the statistical methods are not as clear or reliable and instead described creating a standard curve from a reference gene. There was more to it than that, but I don’t remember it fully/didn’t understand it fully at the time. But, his method makes the most sense of any that I’ve heard some far. I hope that’s something that is talked about a lot in the next few years. Because coming up with a standard way for dealing with data in order to ensure its validity is an important thing in science.

Most exciting of all, some students from the Moczek lab, where Emilie did a post-doc, stopped by. The pointed at a few ghosts of regression lines in my insulin plots. I did find any correlations between insulin genes and doublesex expression in my data as I predicted. I thought this might indicate a relationship between insulin signaling and the doublesex pathway and it would at least have been confirmation that maybe it was the nutritional manipulation that was affecting doublesex. But, they pointed out that in some of the graphs, the alignment of the data points could maybe show a relationship. This is where you start to use words like “preliminary results” and “further investigation” but they were right! I hadn’t seen those maybe-relationships before. Their point was that maybe if I had had larger sample sizes, those correlations would be significant. Of course, maybe they still wouldn’t be, but it was still exciting to hear that. Especially from people that know a whole lot more about insulin and bug development than I do.