The past few weeks have been intense. I’ve spent so many hours reading, watching, and doing whatever I can to learn more about neural nets. In my first post, I explained the basics of machine learning, and in my second post, I explained what a neural net actually looks like. Now it’s time to delve into what actually allows a neural net learn from its data.
This post is coming to you a little later than expected, but hopefully I can get you caught up on my last days in lab this summer. I spent a lot of time on excel, analyzing the data collected from the video analysis. If you remember from my last post, I took lots of videos of moving sperm from swm-1 and spe-6(hc187) mutants in order to compare how SPE-6 affected the motility of sperm that had activated precociously. While it meant a lot of time on the computer, I still really enjoyed it because it meant that I got to see an experiment through from beginning to end (with help from several lab members). I spent most of my spring semester and portions of my time this summer getting trained, so it was exciting to finally feel like I had completed an experiment fully.
Throughout most of his career, Gardner struggles with both the idea of all people being able to reach their potential and the common good. He talks about the importance of individual potential and the common good and the barriers to both. In his early writings, he is ambiguous about how society can both ensure individuals can reach their potential and focus on the common good when oftentimes the common good can inhibit individual potential. In the end, he settles on strong leadership as the best way to achieve both ends.
In my last post, I mentioned the process by which nematode spermatids activate to form spermatozoa. Sperm activation in nematodes is characterized, among other changes, by the formation of a pseudopod. Whereas vertebrate sperm “swim” using a flagellum, C. elegans sperm “crawl” using a pseudopod. The goal of both is to locomote towards and ultimately fertilize an oocyte. In C. elegans, the process by which male sperm activate is tightly regulated because successful fertilization requires sperm to activate at precisely the right moment – when the male sperm are transferred to the reproductive tract of the hermaphrodite (Stanfield and Villenueve, 2006). If sperm activate precociously (within the male), then they are often unable to be transferred to a hermaphrodite. Many proteins control this process of sperm activation to ensure that spermatids activate properly. One such protein is SWM-1, a protein that functions outside of sperm to inhibit sperm activation in males prior to insemination of the hermaphrodites. In swm-1 mutants, however, sperm activate precociously and often fail to be transferred to a hermaphrodite. This precocious activation phenotype extends to other mutants as well, including spe-6 mutants. In spe-6(hc187), the sperm of males raised at 25°C lack SPE-6 protein and activate precociously within the seminal vesicle.