Aloha everyone, first I wanted to say everyone who helped with the cookie break on Monday, nice job. I thought you guys did excellent in making the cookies taste and look good. This week I did some more interviews and can start working on telling the story of how the HABs forecast model developed. My survey is getting great results, thanks to all of you who participated and shared. Sorry I tricked you with the lure of free food when there was none. I am still having trouble getting males to respond to the survey. WHATS UP WITH THAT GUYS!!!!! I thought the seminar with Lauren was awesome and in addition to the q&a session we had with her in the Pacific Forum, Emma and I had a very nice lunch with her and a few other MBARI peeps. I also had a few meetings with my mentor and Fred about the plan for open house. I have been working on some augmented reality stuff for that too but so far the website is all screwy and its not working well. I hope to have this resolved beforehand. If the website doesnt get it together in time, I am thinking of a backup plan. I am really looking forward to kayaking on Monday. Not looking forward to running out of time. I have so much to do and am very optimistic I will get it all done
Mahalo for reading
Alia

Hello everyone!

This has been a particularly productive week for me.

First of all, had a great weekend diving with George and Diego (I'm going to write a blog post about it, stay tuned!), sleeping under the sardine-nado at the MBAq, and admiring blue whales. I definitely recommend taking the opportunity to go whale watching for free. Also, thanks to everyone who tried my vegan chocolate chunk pretzel cookies and told me they were good.

I'd say that I've gone from, like, 0 to 50 this week. I was feeling a bit discouraged from spending so much time learning about machine learning and having nothing to show for it, until yesterday when the script I had started late last week was able to load some data and use a machine learning algorithm to recognize clusters in the data. HUGE step forward. Now I am finally doing some programming and will be spending the remainder of the week (and maybe the remainder of the summer, we'll see how this goes...) adding onto my script so it can do more machine learning things.

I've also been working on a STOQS graphic (STOQS being the data visualization software that my mentor created) for the STOQS webpage, and so we can get some STOQS merchandise, namely coffee mugs.

Have a lovely weekend!

When the stars align...

...or in my case- DNA sequences! The mystery has been solved- the software I'm using to clean the sequences was incorrectly assigning the forward and reverse reads, which was making the cleaned sequences wonky. I spent the earlier part of this week re-aligning the sequences and cleaning them up. I'm thrilled that my sequences are now falling out with Stephanie's- HUGE relief!

The rest of this week, I've spent working on the line drawings for our species description and looking up Latin and Greek words to come up with a species name for our Pseudothecosome. I was hoping for something like "mustached dancer" or "tumbling mustache," but mustache in Greek is mytax and doesn't blend well with other words easily. I'll keep looking... You may be asking yourself "why mustache and dancing/tumbling?" Stop by my desk in the midwater lab and I'll show you some cool photos and videos!

If that news wasn't good enough for you- I'm going on the Western Flyer cruise next week! We're leaving Tuesday evening and getting back the following Tuesday night. I'm so excited to go out and spend time learning about the variety of projects going in the lab. We're also hoping to collect some more of our species so I can get a good look at them for the more detailed line drawings.

Cheers,
Megan

Week 5:

I spent this week adding live data visualization functionality to the benchtop simulator software. It was a fun rabbit hole to dive down, found a open source software library and made it work well for my purposes with some tweaks to improve performance. Learned about threading which is a programming technique for increasing responsiveness in programs by splitting asynchronous tasks into separate threads that the different cores of the CPU can run simultaneously. The benchtop simulator software is now really finalized and finished. The next 5 weeks will focus on the new project which we're calling the Inertial Current Profiler or ICP (which is an unfortunate acronym because of that weird music group). I also need to start putting together my presentation and continue work on my project report.

-Pierre

Hello everyone,

I'm glad to report that nothing in the Brewer lab is broken yet.  After crunching some numbers in Excel, I looked at pure water's the change in enthalpy versus pressure and saw that the slope is nearly zero.  In other words, pressure doesn't have a strong effect on the total heat content of a pure water system.  After seeing this, my mentor suggested I ramp up the pressure to 3000psi (which simulates a depth of roughly 2000m); however, the instrument that was created here by MBARI engineers to do that has a limit of 1500psi.  My mentor just said to ignore it (the worst that would happen is an O-ring pops off and water leaks), made me question what the purpose of science is if you can't have a little fun, and said that a different mentor would help clean it up if it did leak.  After spending almost all of Monday in the dark collecting these spectra at high pressures and praying for strong O-ring seals, we did not have any (visible) leaks! I'm currently processing that spectra and I'll be taking the instrument apart at some point to ensure it's working properly.  More to follow later.

 -David

Hey guys!

I have spent this week working on the same thing I was doing last week, which is going through the VARS time-series images from Station M and counting all of the Macruids that are present. Once I get through a full pulse I am going to start planning the type of analysis I want to do in terms of statistical tests and what not. I will be comparing the fish counts from the pulses I get through with oxygen data and climate indices. I am really interested in seeing what the results are going to be. Hope everyone is having a great week!

• Danielle

Hi friends,

Busy busy busy, but somehow there's still a million things that need to be done! Oh well, that's the way the midpart of projects always goes—I'm making some dents and feeling optimistic that I'll get there. (: I got several blog posts up (yay!) and all the intern profiles are in, so I'll be putting the last few up soon! In the meantime, I've been working on designing the ESP page layout as we come up with new ways to dice the different topics and represent them. I have started writing, although there's still a lot of copy and pasting information around. I'm meeting with Jim and Nancy today to go over my progress and figure out next steps/which pages should be prioritized and done first. Tomorrow I'm going to dig in and really work on Dr. Benoit-Bird's website.

DON'T FORGET TO GUESS: http://www.mbari.org/animal-plant-mineral-or-man-made-3/
This one might be a little easier than the last two, so try to be as specific as possible for a chance to win!

Yesterday with Lauren was GREAT! We went to lunch and I had a chance to asking her a bunch of questions, plus then we had our intern forum and I got ask even more. I feel like she gave me some really helpful advice and I was super excited to learn about the ShareScience program. A database of friendly and open to interviews scientists is SO valuable to science journalists and I'm really glad to see that groups are working to foster better relations between scientists and media. It was also really neat to see that she had a microbiology background like me but is now working in earth science (so there's hope for this bacteria nerd yet) and I loved talking to her about her career pathway. Also, I'm back on Twitter/Instagram game so keep an eye out for some fire emoji posts.

Also had a lot of fun whale watching this week. Saw a BLUE WHALE (100% worth the subsequent sea sickness), and then went straight to the aquarium sleepover which was great. Shout-out to Ellie for making like 18 dozen cookies and only cutting me off after I'd eaten probably a dozen.

Emma

P.S. If I look really stiff and uncomfortable (and/or smell like Bengay or Tiger Balm), it's because I accidentally threw out my back in a minor way. No medical emergency, just a bit sore! (:

Greetings!

Wow I can't believe we only have 4 weeks left...I feel like I only just got started on my project, and I'm hoping I can finish it in time! Aah!

This week has been super fun and I've learned how to do a lot of things that I only barely remember from college lab courses. Last Tuesday I made cDNA from transcriptome RNA that was extracted from vampire squid arm tips (we used the tips because we know there is bioluminescence there, and is the most likely place to get the right sequence). I spent the rest of last week designing primers so that I could amplify what we hope is the gene for the vampire squid photoprotein. In addition to making primers that had the right melting temperatures/GC content/structure, I thought that I had to make unique primers that would exclusively amplify the photoprotein gene, so I spent a lot of time blasting my primers against a database to try and find ones that wouldn't complement any other sequences from our transcriptome. Turns out I don't really need to be that specific for the amplification, so I ordered a ton of not-quite-as-specific primers on Monday, and those came yesterday. So I ran a PCR yesterday afternoon (sorry I missed the Q&A), for three different gene sequences that might be the photoprotein and I'm running those amplifications on a gel today to see if we amplified the correct sequences. I'll find out in approximately 20 minutes. Keep your fingers crossed y'all.

If my gel bands look like they're the right length, I'll cut the bands out of the gel, purify the DNA out of the agarose, and sequence the purified DNA to see if it matches the sequence from the transcriptome (the one I designed my primers from). If it looks right, I'll clone that sequence into an expression vector and try to express it to confirm bioluminescence. Whee!

Happy thursday!

 Chloe

Another week goes by...

And progress is being made. This week I finalized the content of my first set of analyses (observations, lagged cross-covariance, daily and upwelling-scale band-pass filters, seasonality of daily pH) of the may-december 2015 time series of OA1, OA2 and M1. I made most figures presentable and formatted the introduction and methods sections to fit our final paper's guidelines.

Yesterday I began working on expanding these analyses to the most recent time period where data is available for all three moorings (october 2016-july 2017), which is exciting as I get to test whether my conclusions hold during a different time of the year (unlikely for most) and in this most recent, very wet winter and spring.

I'm also thrilled to go on a short maintenance dive of OA1 tomorrow aboard the Paragon. I'll let you know how that goes next week.

Happy foggy days,

Diego

Hi Everyone!

 This week has flown by crazy fast. I spent most of Monday extracting my data from VARS and finally getting everything into Excel sheets to start analyzing! Also setting up the cookie break with Ellie and David (I had an awesome time doing that you guys! Tuesday was a lot of the same thing, Danielle helped me out a bit and I was able to find a better way of organizing and calculating all my data. Spent most of yesterday either talking with my mentor or at the seminar/with the seminar speaker. Today so far has been extracting my length measurements from VARS and sorting everything to the right individual. It's been really cool seeing some correlations and how my hypotheses have been so far!

Other than that I've also been working on an abstract for SACNAS with the REU, and applying for the travel scholarship as well. 

Hope everyone's weekend is amazing!

Victoria 

Hello!

I have started making progress on the pre-processing of these spectrograms. There is a lot of great code on github from people who have similar sound analysis work on whales. Danelle was awesome and kind enough to push that onto the repository for me so I could start manipulating it to fit our specific whale call parameters. I have also been looking into specific CNNs that might help me get a better idea of how the architecture of our CNN will turn out.

I have also dedicated time into the SACNAS application that is due this Friday. Just got back abstract revisions from both of my mentors and will be submitting it today. It seemed odd to apply for a presentation that I am ready for yet so we'll how it goes.

The cookie break was awesome! I appreciate all the work you guys put in to making them.

Enjoy the weekend,

Daniel

Week 6:

At this point the SeeStar (RPi) is to operate as a dictator to the Feather M0, while the Feather M0 acts as an intermediate processor between the SeeStar and its sensors. The goal is to have the SeeStar simply order the Feather M0 to command sensors to record a sample and go back to sleep! 

Thus far, I have four serials that can read and store data onto an SD card with three that can serve as means for communicating to sensors and one reserved to talk to the SeeStar. I experimented a bit and daisy chained the pins so that they all talk to their subsequent serial. I also mapped out which pins belong to which SERCOM along with its associated keywords and pads (syntax and assignments were big issues).  

 I also got to the point where the Feather wakes up after a user presses a key and then goes back to sleep. Essentially, that is the main goal of the project, but I am concerned about how the Feather discerns this wake up call. I have to make sure that the Feather wakes up from a serial signal not just a digital one. 

Hi Everyone!

This week I was able to make great progress on my research, yay!
I'm hoping I'll be able to complete my project by the end of the internship (fingers crossed). My mentor and I are currently working on another algorithm that can detect the max chl. concentration in the pct data which includes chl, temp, sal, O2, and par. We're looking for seasonal variability and daily variability as well. The programming part of the project is coming to an end and the data analysis is to begin (so exciting!). Also in the works this week: my SACNAS application! (deadline tomorrow)I've submitted the abstract earlier this week following Francisco and Monique's edits and approval. currently working on the travel scholarship application, it requires a recommendation letter, CV, and personal statement.

Thank you guys for the delicious cookies on Monday! I'm looking forward to round 2 of kayaking in the slough next week!
have a great rest of your week

-Ash

Hi all!

I had a great weekend with pie making, whale watching (2 blue whales!!), aquarium sleeping, and huge amounts of cookie baking! Thank you so much to everyone who baked cookies and made the first intern cookie break a success! We had seven different types of cookies and banana bread.

Most of my day-to-day activities remain the same. I sample and run tests on my cultures in the mornings and work on graphing and analyzing data in the afternoons. Growing my cultures under seven different light levels at the same time has been working really well! We're seeing patterns emerging in growth rates and fluorescence. Next week we're going to do more intensive photochemical analysis on the cultures to study how their photophysiology changes at the varying light levels.

Unfortunately we've had some setbacks in the chemostats. We've had problems with the pH falling in the chambers, and could not figure out why. We had been continuing with the experiment despite this, and and this weekend we extensively sampled the chemostats to get a baseline for the next phase of the experiment. On Monday, my mentor decided we needed to take a step back and figure out why the pH was decreasing and the growth rates were lower than what we have been seeing in the incubator cultures. So, we ran a bacterial test and tried changing some settings on the chemostats. Our suspicions from last week were sadly valid and two of the four chemostats have been contaminated.
As of yesterday things are looking better though! We're now giving the chemostats antibiotics every couple days, which don't seem to be hurting the algae at all (yay!). Lowering the amount of air bubbling into the tanks seems to have solved our pH problem, so we're going to go ahead with the CO2 and NO3 experiment in the next couple of days!

My family is visiting this week, so I'm going on another whale watch on Sunday!! Hope everyone has a good weekend!

Hey everyone!

To complete my project, I decided to to back to an approach that I used earlier this summer. I manually annotated (drew boxes on) around 600 images. These images will be used as training data for my neural network. After I train the neural network on this data over the weekend, I will use the neural network to automatically create annotations on new images. I will then go into the iterative cycle of correcting these predicted annotations, and integrating them back into my training data.

For my new training data, I revamped my directory structure and re-wrote a lot of the code. It is really great to see how much better I've become at managing data and code-reuse while working on three different machines.

I also did some more work on the effects of interlaced video on convolutional networks. I put my code in a loop so I could see a bigger distribution of results.

Nathan