Helen Krueger, undergraduate student
1 February 2020
My project has hit a momentary lull while I secure the lab space, training, and instruments needed to complete the next phase of my project: processing biological samples for SEM and phenotypic taxonomic identification. I am working with several people in IWS and SciTech to help me with this stage. I hope to be up and running again by Feb. 3rd.
Since there is not much to update on the project front, I wanted to share a semi-new hobby that I have taken up: birding. Going into the environmental science field, I have been told that the number one best thing to enhance your observational skills is to become a birder. Because the animals are so small and move so quickly, they present an observational challenge in being swift and accurately identifying them.
The applicability of birding in your everyday practice of walking to campus, hiking, or even being a passenger in a car, is in the novelty of species that you weren’t aware of existing around you. The main motivation for learning how to be a birder is being outside and cracking open another layer in the experience of your daily walk.
So far, I have been focusing on identifying waterfowl, because they are seasonally present in the winter, and somewhat easier to identify. I have found quite a bit of success identifying birds with the help of knowledgeable friends, bird apps and ID books. All of the birds that I have identified and counted were recorded in ebird - a citizen science birding platform that collects data on birds worldwide. If anyone would like to go birding with me, let me know!
Thankful for undergrads!
Grace Freeman, graduate student
1 February 2020
With the new quarter comes new adventures! It started slowly as we had a week of snow days after only a few days of classes. Even coming from Minnesota, I had never had school cancelled due to snow! Maybe for ice or temps in the double digits below freezing, but never snow. The whole thing was wild and novel to me!
Now that I’ve survived the “massive” snowstorm of 2020, I’ve turned my attention back to data organization and analysis. For my project on individual variability and foraging behavior in the Salish Sea, it’s pretty darn important that I have data on individuals and their foraging behavior in the Salish Sea! To collect this, my team has been analyzing photos taken from Whatcom Creek, and working to match them to other photos taken at the creek in previous years in order to identify the individual seal in each photo. As you might imagine, it’s a slow process. There are some computer programs designed to run this kind of analysis automatically, but we have never been able to find one that works for the dataset. Instead, it’s done by hand.
To give you an idea of just how slow the work is, here’s a quick description of the process:
- Upload photos
- Crop photos to include only the individual to be identified
- Look for photos from a similar time that may contain the same individual
- Look for a front, left, and right angel of the seal in question – if we should be so lucky!
- If there are these alternative angles, hope that are all clear enough to see patterns on the individual’s head and face
- Study the photos looking for markings, scars, or other distinguishing characteristics
- Begin scanning through the database of seals previously identified at the creek
- Compare the unknown individual to each of these seals and hope for a match!
Because there are 150+ seals with IDs in our database, this stage of comparison can take anywhere from 20 minutes to 3 hours for just one photo. Sometimes, you’ll get lucky and be able to identify a seal and then apply that ID to other photos containing the same seal from the same day. Other times, you might scan through the database two or three times without any connections. Is it possible it’s a new seal? Yes, but, based on knowledge gleaned from the project thus far, it’s much more likely that you missed something or it’s an angle of that seal we didn’t have a photo of before.
As you can see, it’s a long and tedious process, but because it is absolutely critical to my project, it’s one that has to be done. I am currently working on a proposal to receive funding so I can spend my summer doing this and still pay rent. For now, I am incredibly lucky to be working with a team of undergrads (headed up by Delaney!) who commit hours of their time each week to sift through these photos and make the connections necessary for me to run any kind of analysis. It’s not an exaggeration to say that I couldn’t do this without them!
Bobbie Buzzell, graduate student
1 February 2020
At the beginning of the month, I was able to finally finish cleaning scats! I have met a great benchmark in the project. While I was able to complete this part primarily on my own, I was happy to have Sehome High School student Ms. Orly Lindner help out on a couple of sessions during last quarter and winter break. Thanks, Orly! You made the damp, dark cleaning room a little livelier. We (myself and my undergrad assistants) are taking on the next part of the project, sorting out the fish bones.
Depending on how you look at it, this quarter started off without a bang. Almost the entire second week of classes were cancelled due to heavy snow in Bellingham. What would normally have been an exciting week of no classes when I was a high schooler or undergrad was a whole week of setbacks for sorting (this grad wants to graduate on time!). Campus closures means no biology building access for myself and my assistants. As disappointing as this was, we were quick to hop back into the flow of lab time. We have reached the latter half of the first 100 samples, which is still good progress given the campus closures.
I’m happy with this graph but I was also interested in how this compares to different predators. I chose the sea lion group, since they are also pinnipeds and are abundant in the study area.
By the end of February, my hope is to have at least 200 samples ready to be sent for fish ID with Mr. Walker, the food habits specialist. Until then, myself and my assistants will keep sorting.
Nathan Guilford, graduate student
1 February 2020
After receiving the final sequence data from the sequencing facility, this month has been spent toying with various programs for identifying a large and highly informative panel of SNPs. Not surprisingly, working with fecal DNA has its challenges, but there are many options for identification parameters and filters that I am using to find the balance between number of SNPs identified, and quality. When looking for high quality SNPs, I am trying to ensure that the loci are correctly identified with no sequencing errors, and each locus is present in at least the majority of the samples. This allows the genotypes to be more easily constructed and compared. With the degraded DNA in scat, I am having some trouble finding a lot of SNPs that are present in most of my samples, but I am continuing to compare methods and workflows.
I am also now investigating what method we should use to construct genotypes and compare them between samples, which will hopefully show us the three identical samples from the aquarium and be used for future use in identifying wild resamples. A promising program comes from a previous student in our lab, Andrew Rothstein, who developed WISEPAIR.py (Rothstein 2015). Using PCR amplifications of microsatellites, Andrew created a script that makes a frequency distribution of all pairwise comparisons of a set of samples for the product lengths (and therefore alleles), and calculates a threshold for which pairs of samples are either not similar, similar due to chance/errors, or similar and likely resamples. I feel like this method would be perfect for scaling up to some wild population analyses, so I am hoping that it can be adapted well for the SNP data that I will be collecting.
On another note, this month I was invited to attend the Seattle Aquarium’s alumni event, where previous volunteers/employees were able to present the work/research they have moved on to perform. This gave me the chance to see Hogan, the harbor seal who provided us with fecal samples for our genotyping controls. As you can see, he was very excited to hear about the project!
- Rothstein, A.P. (2015) Non-invasive genetic tracking of harbor seals (Phoca vitulina) (Master’s thesis). Western Washington University, Bellingham, WA.
Delaney Adams, undergraduate student
2 February 2020
The start of the winter has been a rainy one for sure. I’ve been keeping busy with sports and all of my classes, as well as with work and preparation for post-graduation. Along with all of that, I have spent a large portion of my lab time this month working on fixing one of our old codes for making new observation schedules. After quite a bit of time spent working on that, I ran into many issues that I couldn’t solve on my own, and summoned the help of two of the other grad students: Grace and Jonathan. They were both an incredible help, and after many long hours, I think we finally have it mostly solved.
Later this month I finally had time to start digging into my project, and that involved just rummaging and sorting through the data. Through my first dig through the data, I found a couple of interesting facts, but my I still have much more work to do in terms of exploring the data since not all the events were independent of each other. Next month I’m looking forward to collaborating with Madi, the manager for the project last year, in order to fully understand all of the things she did and found. I’m also looking forward to collaborating with Grace to hopefully use some of my newly learned GIS skills to make a map that explains some of the confusing parts of her project. Lastly, I’m looking forward to digging deeper into my data and seeing what I can find.
Jonathan Blubaugh, graduate student
2 February 2020
I had my committee meeting this month and got approval to finish my work and write up my thesis. I’m very excited to be on the final stretch, even if that stretch is 4 months long. I’ve signed up at the Writing Center on campus for a thesis partner to keep me accountable for my writing progress. I’m planning on taking one month per section of my thesis and then an extra month for final editing and formatting. If I keep that timeline, I should be able to defend in mid-June.
One thing that came from my committee meeting was using a different multivariate approach to visualize my data. I’ve been working on a Constrained Ordination, which is similar to the PCA shown in my last blog post. The difference is that instead of using the output of the model as the axes, I can force one axis to be the sex ratio of each model output. This means I can show the amount of variation explained solely by the sex ratio changes on one axis and the other axis is the variation explained by changes in the impact. This will hopefully provide a more robust analysis of the differences between the harbor seal sexes. I’m continuing to work on this analysis during next month but will hopefully be done with it by the end of February so I can write my results up in March.