Author Archives: Ricardo Matias

I am very grateful to have been given the opportunity to participate in BSURF this summer. As a person that did not have any research experience going into the program, I am pleased by how much I have grown as a scientist and as a problem solver. I believe the most valuable aspect to me was learning about how science works, and how knowledge is produced. I’m also extremely happy with the relationships I formed with my other peers in lab, and in BSURF. Fortunately, I will be able to continue to work in my lab throughout the semester, and am eager to continue to learn, improve, and ask questions.

I also wanted to comment on how much I enjoyed the weekly seminars. Having scientists come in from several departments at Duke to introduce us to their research and to tell us about their path through science was incredible. It really showed me that you never know what you might end up doing with your life, no matter how set on a goal you are.

Lastly, I wanted to thank Dr. Grunwald, and Jason Long, for their work in running the program, and Trinity College for funding the program. It would not have been possible without your support!

I thoroughly enjoyed the seminar portion of this program. It was valuable to me to have scientists come in, talk to us about their path through science, and introduce us to their research. It was great to hear about how some ended up in the field accidentally, while others were committed to science from the start.

One of my favorite seminars was delivered by Dr. Anne Yoder, director of the Duke Lemur Center. As someone working in a biomedical engineering lab, it was awesome to hear about about a completely different field of research. It was interesting to learn that the first lemurs that arrived in Madagascar likely drifted to shore on a “raft” from mainland Africa. From there, evolutionary mechanisms generated the diverse range of lemurs we observe today. Visiting the Duke Lemur Center following this seminar reinforced these concepts. I hope that one day, I can be as passionate about a field as Dr. Yoder is for her lemurs.

With two weeks left until the poster session, my project is finally starting to come together. I will definitely be working overtime to make sure that everything is completed in time.

As far as my project goes, I’ve improved quite significantly at using MATLAB, so things are going much more smoothly. I still find that, due to my inexperience, I lack necessary foresight to avoid making mistakes. At times, it feels like after taking two steps forward, I take one step back. Although this can be frustrating, it truly is the best way to become familiar with the programming language, and the field itself. This foresight only comes with experience, as evidenced by the progress I’ve made in just six weeks.

In addition to working with MATLAB, for the past few weeks I have been reading through neurobiology and neuroengineering textbooks, to obtain a deeper understanding of the evolution and fundamentals of the field. It’s interesting to see how improvements in technology have paralleled our ability to record activity from the brain, exemplifying the idea that many fields of science inform and develop each other.

In my opinion, one of the most interesting chalk talks last week was Emre’s discussion on the Foxj1 protein, and its role in ependymal cells. I think it’s fascinating that the entire differentiation process of a cell can depend on solely one transcription factor. Our body contains millions of cell types, yet the mechanism that distinguishes a liver cell from a neuron from a skin cell can differ by just one protein. Emre is trying to understand why and how these proteins degrade so quickly, even though they are so important for ependymal cells. To do this, Emre is mutating various sites on the protein. After finding which site is responsible for degradation, Emre’s lab will use that mutated protein to overexpress Foxj1 in ependymal cells and observe how they react. This information can be crucial in fighting certain neurological diseases. Emre’s project was also particularly interesting to me, as my lab is using a completely different approach to combat neurological diseases.

My day in the lab begins at around 10 am. As I mentioned in my week one blog post, my day typically revolves around trial and error. Although I am getting much better at using MATLAB, I continue to refer to online documentation, or textbooks that my secondary mentor provided to me, in order to properly use commands.

After starting up MATLAB, I usually just pick up where I left off the day before. I always have my lab notebook present in case I need to organize my thoughts, or plan my approach. Another technique I use – and recommend – is to think about code in the most rudimentary steps. Although computers are extremely sophisticated machines, they cannot predict our thoughts. I’m constantly thinking to myself, “ I need to take that variable from over there, manipulate it, and place it over here.” This thought process has really helped me understand how MATLAB, and code, works.

Whenever I complete a small segment of my project, I touch base with my secondary mentor to make sure I am going on the right track. I also speak to him whenever I think of another possible method to accomplish a desired goal. Although I am working with code, I still try to challenge myself by thinking of questions or different methods to achieve a task. It’s also important for me to never lose sight of the end goal, whenever I feel sidetracked or stuck with a segment of code. I leave lab anywhere between 5 pm and 7 pm, and I always make a note to myself in my lab notebook about where I left off.

The Nicolelis Lab is well known for researching brain-machine interfaces (BMI) in an effort to develop brain controlled prosthetic limbs to be used by patients suffering from quadriplegia. Although the lab has multiple ongoing projects, my research falls within the BMI project.

For my project, I am using data that has already been collected by the lab. Using this data, I am calculating lag times between velocity models and neural firing rate models. My hypothesis is that the lag time will increase when a monkey is completing multiple tasks simultaneously, rather than just one task. Knowing and understanding these lag times is integral to the application and construction of brain-machine interfaces to be used by humans.

Although I am unable to go into greater detail of my project on this blog, due to the sensitive nature of my data, I will be sure to edit this blog entry after the data has been published.

filename = ‘blog1.mat’;

As someone with limited exposure to programming and coding, knowing that I’d be working with MATLAB for eight weeks worried me. I wanted to finish B-SURF with a greater understanding about how science is produced, and I was afraid that my inexperience would hinder my progress. Interestingly, MATLAB has already taught me some of my most important lessons in science.

1. Learn by trial and error: This is basically how I’ve learned everything. Whenever I need to execute a function, I search for the type of command I want on mathworks.com/help/matlab/index.html, and experiment until it works. Watching an error message disappear is one of the most satisfying feelings.
2. Collaboration is important: Luckily for me, some of the other undergraduates in the lab have several years of programming experience. Working with them is always extremely helpful. I spent most of the other day trying to optimize a script, and one of the undergraduates rewrote the entire code in 10 minutes. Thanks, Alan.
3. Science can take on many forms: Throughout high school I always held a notion that scientific research was a bunch of people in white lab coats watching over boiling, colorful liquids or looking through a microscope. Little did I know that my first research experience would involve a lot of MATLAB, and few white lab coats.

I’m much more optimistic at the end of week one, and I didn’t expect to enjoy my work as much as I do. I’m excited for the work I’m doing, and am eager to see where my project goes.

save(‘blog1.mat’)

clc