Happy basically Friyay everyone! So I don't have much to update on. Yesterday was spent working on a protocol for my chaetognath time sequence experiment I'm going to be conducting, and then after I completed the protocol I sent it to my lab for some feedback. I then took their comments and edited my protocol, and at this point I'm just trying to figure out some specific volumes to use. For a little background info, the time sequence experiment is just something I'm doing so I can see the rate of decrease of the light production of chaetognaths over time. Might be useful information to know in a paper on cheatos
Safe travels over the holiday everyone!

Hey guys! Happy 4/7/18 week.

Another long week at MBARI. Yesterday was spent finishing off the last of my extractions! Finished them off just in time to make it to the intern BBQ, which was a lot of fun getting to know people from all the surrounding institutions.

Today will be a day of experimenting with qPCR, getting my standards correct and reliable before doing the real thing after the break.

IMPORTANT NOTICE: ENGLAND IS PLAYING COLOMBIA IN THE WORLD CUP AT 11:00AM - GUARANTEED TO BE A THRILLING GAME WITH MAYBE ONE OR TWO GOALS (IF YOU'RE LUCKY). TUNING IN IS CONSIDERED MANDATORY.

Have a great break!

Hello everyone! Since last Thursday I was able to get my p-values and all five of them were significant which is super exciting!! I was hoping for just one significant value but five is even better! I also made scatter plots for my paper to show the relationships of my variables. I have been working on writing my methods which has been a bit of a learning process because I have never written a paper where I had to cite codes and coding software. I am currently reading papers and figuring out how all the measurements of the data I used were taken for my methods. I have started my outline for my open house poster and hope to finish that next week. Happy 4th of July!

happy holiday everyone!

This past week has been more problem-solving. Turns out one of my candidate luciferase sequences is actually paramyosin. Which means it is definitely not the luciferase. So I am working with different free software programs to try and blast the raw MassSpec data from the luciferase protein isolation with the transcriptome database to try and find new candidate genes. I'll also be going up to the Mass Spec at UCSC for some training, and we can do a more detailed analysis of my data then. If that all fails, I will basically have to start my project from the beginning; redo the MassSpec from new protein purification, find new candidate genes, and try to clone and express those!

So I'll need all the luck I can get. Have a good long weekend everyone!

Hi everyone! even though it has been a short week, I have gotten a good amount of work done. I finished my larvacean molds and received 3D prints today, they look great! they will definitely be helpful when I try to produce as many tails with different properties in the upcoming week. I have also finished my tank setup, and I have my camera, plates, and physical model setup so things will be kept consistent in testing. Additionally, I have started writing my Matlab program for analysis of the data. I will be working on analyzing the trials mathematically. I am also still doing research on coiled nylon to reproduce artificial muscles.

Hi everyone,

Since last week I've continued analyzing coral site marker footage, noting changes in the size, health, and epibiont communities of corals over time. Yesterday I transferred abiotic variables into my epibiont spreadsheet, so I should be able to begin proper analysis as soon as I finish going through the remainder of the site-marker footage. Today I plan to continue working through footage, offer my priceless 21-year-old wisdom to the students at the wetlands watch tour, and learn a bit more about my responsibilities on the Western Flyer next week.

Hope everyone has a great holiday!

FRIDAY: Attempted to re-organize my code into a standard Python package. It was a total coding failure. Broke everything within two hours. I am thankful GitHub exists.
MONDAY: Struggled (and finally succeeded) with Python package creation.
TUESDAY: More coding.

Hello everyone!

This week was very short, but it was still very eventful. I worked on implementing a new method to detect optimal change points due to data sensors drifting on a profiling float, without compromising or removing actual data that is not due to drift. This method naturally still needs a lot of work, and I have to research more in depth on how accurate it is, as well as establish guidelines to follow when using it. This will most likely be my primary focus for the rest of this week, but the ultimate goal is to establish a very reliable method to best correct incoming profiler data, without any bias or removal of real, potentially important data.

I am still excited to keep working on this, and very happy I will be able to continue working on it from home on my off days. Hopefully everyone else's projects are also coming along well.

Have a great trip, everyone, and safe travels!

This week I'm feeling the effects of an empty MBARI. I've come into a number of roadblocks related to assessing the success of algorithms in estimating carbon system parameters in Monterey Bay. The biggest problem can be chocked up to a classic 'dirty data' issue, but everyone who knows the secrets of the database I'm using is out enjoying the holiday. Nevertheless, I'll be going to a sparsely-attended lab meeting with questions in the hopes that my mentor & co. can offer some insights on how to move forward. So today I'll be spending the afternoon making sure I understand exactly the problems I'm facing so that I can get the best ideas towards solutions. Happy 4th everyone!

This week, I've been working on some simple data analysis for my mentor.

I spent the last two days trying to learn how to despike data. I'm sure there are lots of ways to do this, but the method I'm going with is the median filter. It's a straightforward way of looking at a collection of spikey data and then despiking it. Basically, if you assume that any set of data will have radical outliers, you take a subset of the data, say 3 points, and then pitch the highest and the lowest and putting the median into a new data set. Systematically repeat that with all your data and eventually you will smooth out the spikes. Luckily, python has a library in the scipy module that has prebuilt median filter.

I'm also working on rewriting a bunch of his data analysis tools in python. Basically, he has some data that he's collected and analyzed in that scourge of a program matlab. (I loathe matlab. I'm sure I'll be forced to use it at some point, but it drives me insane how much it costs. I'll get off my lefty commie soapbox in a second, but I have no idea how matlab justifies their prices. I mean, I'm fine with paying for some software, but holy hell, the cost of a year's seat license for matlab is just out of control.)

My task is basically to recreate the matlab tools he's used in python and then run another dataset through. The problem is that I don't really know all the things that matlab does so I have to spend an inordinate amount of time reading up on matlab functions and figuring our the corresponding commands in python. If they exist. But it's not like this is just swapping key words for key words here. Matlab and python don't always play in the same way, so I have to figure out the underlying use for the matlab code and then figure out what tool or tools in python will do that thing.

Anyway, that's what I'm working on.

Enjoy your holiday everyone!

Hi everyone!

Since we started the abalone experiment last Wednesday, this week we just have to go through our husbandry routine everyday. This includes feeding the abalone, cleaning/draining their water, checking flow into the jars, and checking for mortality. We also have to make sure the system is running correctly so we take temperature readings of the water and wipe down the pH and O2 sensors.
On Monday we found that 10 abalone had died over the weekend (mostly from the jars exposed to stressful treatments).
Today, we only had one mortality.
I'm going to be coming into the lab on Thursday and Friday as well to continue with our routine.

Happy 4th and extended weekend to everyone else!

Short week but I did find some interesting findings! Based on the songs I've listened to and categorized, I have decided on a frequency range (50-5000 Hz), which will hopefully eliminate any background, non whale noise. Given this frequency range, the pitches range from G1 at 49 Hz to C#8 at 4978 Hz. (G3 is the lowest note on the violin= 196 Hz, A4 is 440 Hz for reference) That's 6 octaves!

For the rest of the day I will continue labeling units on songs as it'll be helpful to have a decent set of labeled data with which to compare with the automated system.

Happy 4th!

Hi everyone,

Last week, listening to the Deep 2020 meetings was fascinating. So much of what they were saying is similar to Life in the Dark and I'm delighted that the exhibition in London covers most of the main messages that were coming out of the discussion. Sitting in a room with Steve Haddock and Bruce Robison talking about bioluminescence was almost as exciting for me as seeing the ROV at work.

For my actual project, I've got my plan agreed for the phytoplankton exhibit, so I'll start developing that over the next couple of weeks ready to test at the open house. And I've also been working on some short narratives for the CeNCOOS data portal to explain the data.

And unrelated but still a big achievement, we successfully hired a car this morning after 9 different attempts/companies/problems. So watch out if you see a bright red Kia drive past!

Greetings and salutations everyone,

On Monday, I reviewing code and planning out with Mayra our next steps after our current part of our project. I ran into some errors trying to integrate Mayra's code with mine. Today, we finished our small part of the project and continuing with the next steps by reviewing notes that were left by our mentors.

We will be reviewing code from the previous project to help us with creating our own in order to display what has been detected in each frame.

Hi everyone!

This week was too short that I forgot to post. This week Raymond and I integrated our code together. We got a lot of errors by some libraries our machines did not support. After all, we successfully were able to open every xml file and extract the x-y positions. Now, I'm initiating the events on the ROV footage and working in a class that will detect the box object.

Happy 4th of July!!!

Hi everyone!

This week was too short that I forgot to post. This week Raymond and I integrated our code together. We got a lot of errors by some libraries our machines did not support. After all, we successfully were able to open every xml file and extract the x-y positions. Now, I'm initiating the events on the ROV footage and working in a class that will detect the box object.

Happy 4th of July!!!

Hello people,
Sorry for the late post. As Larissa posted earlier, we did abalone husbandry this week. We checked their food, mortality, and cleaned their chambers. We than do image analysis on their shells when the husbandry routine is done.

missing a few updates - be prepared to present at Monday's meeting.

week six missing: (Kevin, Thomas, Richard, Ethan)

Sorry for the late post, as I said in the meeting I got the Iterative Closest Point algorithm (ICP) working on the multinbeam sonar data. My implementation seems to get results very close to what I get by using purpose built software so that is a good sign. The software available is intended for point clouds from 3d scanners. My project differs in that it needs to ingest the LiDAR or sonar data, transform it to to a local reference frame, and then align the different swaths of data. Preferibly with an optional user interface, since it is usually can operate in the background and just provide alignment results to the larger software system that is used to analyize data.

 Here is a screenshot of to multibeam swaths being aligned!
 

Hey all,

 Sorry for the late update. I was out the past week, and sick on the Friday before. What I'll be working on in the upcoming week is improving my stereo calibration results while I wait for the external collaborators on the FathomNet project to request visualizations of the image set data.

Hi all,

very late post due to the short week (and forgot). Spent the Monday & Tuesday mostly sketching out some design packages, yesterday I got to meet with Kakani which helped point me in the right direction. We likely are going to implement a stand alone VHF transmitter that can just we can epoxy to the iTAG in order to have it be controlled by its own separate battery source. Cuts out a lot of the hassle of dealing with a power budget/running out of battery issues.