FAQs

FAQs

What is NeuroData?

NeuroData is a collective of individuals dedicated to our mission, and a repository of the data and tools we provide as free and open resources to the community.

What is the Open Connectome Project?

In 2011, we launched Open Connectome Project, which is an open source software stack of Web-services that store, analyze and visualize large imaging datasets. However, as technology changed, features were added, and scale increased, our academic development team and resources became overwhelmed. We overhauled our custom stack into a community-built and maintained software ecosystem deployed in the commercial cloud neurodata.io, integrating multiple open-source projects and extending them for our needs. The ecosystem enables analyses on disparate datasets by re-using components originally designed for other applications.

Q & A

Respondents:

  • Joshua Vogelstein(@jovo) - PI for NeuroData
  • Adam Charles(@Adam) - New BME Faculty
  • Jeremias Sulam(@jere) - BME Faculty
  • Eric Bridgeford (@ebridge2) - 3rd-year Biostat PhD Student in NeuroData lab (JHU undergrad)
  • Jaewon Chung (@j1c) - 2nd-year BME PhD Student in NeuroData lab (also was BME Masters student at JHU)
  • Jayanta Dey (@jdey4) - 2nd-year BME PhD Student in NeuroData lab
  • Ben Pedigo(@BPedigo) - 3rd-year BME PhD Student in NeuroData lab
  • Hayden Helm (@helm) - Former Assistant Research Engineer (JHU undergrad / masters)
  • Thomas Athey(@tathey1) - 3rd-year BME PhD Student in NeuroData lab
Q : How is life as a graduate student at JHU (or life at Baltimore in general)?

A : @BPedigo
I love it! I have been happy with my work/life balance most of the time, and I have enjoyed getting to explore the city. Lots of good food and drink here, IMO. I have also found many close friends within the program and think we have a good community here in the PhD program.

A: @tathey1
It's great. Regarding JHU life, I'd like to point out that as BME students, we have access to both the GSA (student organization at the medical campus) and GRO (student organization at Homewood campus). It is worth checking out their websites to see the events that they host. My favorites include the weekly coffee hour, and the roller skating/bowling event they had this past semester (since I'm talking to prospective grad students so I'll re-clarify that events are free). Secondly, JHU loves to tell prospective undergrads about their sheer number of student clubs (like over 300 or something. This is still relevant to us because grad students can still participate in most of them (hmu if you are a tennis player). I find Baltimore to be affordable, and easy to get around. A decent number of students will end up having a car here, but it is not essential since you can usually find a friend with a car, or you can use the city's free buses. I personally like to look for events here. I can't say that the entertainment is as high caliber as places like NY or SF, but the upside is that you can become part of almost any circle if you want to.

Q : What should I expect as a first year student in regards to class/research balance?

A: BPedigo
I wish I knew quantitatively how much I spent on classes vs. research last year, but if I had to guess I'd say 50/50 (+/- 10%). Both are definitely important, and I think there are a lot of great classes here that have helped me grow a lot and have been helpful for research. Though, I'm definitey here to do research!
A: @jdey4
About the workload, I find it the optimum. It is neither too much nor too less. Don't stress a lot thinking about the workload now. I do a lot of mistakes and people are friendly and helpful. We have weekly meetings where you can also learn from other people and their mistakes.
A: @j1c
Since I was a Masters student at Hopkins, my coursework is a bit lighter than others so I am spending about 70% on research and 30% on classes in the first year.

Q : How much time does TA-ing take?

A: @j1c
The current requirement is that you TA for one semester. I would consider TA-ing to take as much time as taking one full course.

Q : What are some of the current projects the team is working on and what must-know techniques are being used?

A: @Adam
I want to give some info for my lab since I just joined the faculty (lab officially opening it's doors in July). My work is at the intersection of Signal Processing/Machine Learning and Neuroscience. The work I focus on spans from the more instrumentation focused (using computational tools to interpret/process neural recordings as well as designing new, improved techniques that merge novel hardware and algorithms) to the more theoretical (how do we extract interpretable latent states from neural data and how can we understand the theoretical properties of those models). Specific projects I'd love to start immediately include 1) working with a wealth of optical imaging data I have from my collaborators, in particular two-photon calcium imaging, 2) theory work analyzing the capabilities and limitations of recurrent neural network models, with an bent towards systems neuroscience modeling, and 3) dynamical system analysis: how can we learn interpretable, relevant dynamics in neural data. My work is also very collaborative so there's opportunities to work closely with existing and new collaborations with experimental labs as well as other computational labs on campus (Josh, Jeremias, etc.). There's an amazing computational group in this department that I'm excited to be a part of!

Q : Are projects more theoretically driven or empirically driven?

A: @jere
I think there's a very large spectrum of projects, both from very applied to very theoretical, as well as spread over different topics and fields of study. In my group we have projects that are very much on foundations of machine learning, and thus theoretical, as well as applied project on a few biomedical applications on brain imaging and interpretable analysis of microscopy images. Several other faculty also span both applied and theoretical problems (Jovo's, Adam's, Rene Vidals, and others).
A: @Adam
To add to this, this spread is what drew me to jhu over other opportunities. There's definitely appreciation for both and plenty of faculty that span that range. I'll also use this moment to point out that I found this a rare quality when I was on the faculty job market that sets jhu apart from other bme departments. It was hard to find a single department that really appreciated both aspects so completely, and jhu bme definitely does.

Q : How do students get exposed to data acquisition techniques?

A: @jovo
Any student working on data i tend to send to the data acquisition institutions/labs for as long as neccessary, typically 1-2 weeks at a time, but can be months as appropriate. the better we understand the data acquisition, the better analysis we can perform.
A: @Adam
This definitely depends on the type of project. Computational labs often work with large amounts of data collected from experimental labs. Larger projects will often have this flavor as the sheer scope of the project necessitates a higher level of compartmentalization. Smaller projects can be much more intertwined, especially when creating computational imaging solutions. I'll let others speak for their labs, however I'm actively starting up a few collaborations with the possibility of grad students visiting and working in the experimental lab for a few weeks (for example motor action in monkeys and reward representation in mouse hippocampus for two) to have my students get a better feel for where the data comes from, and all the all-too-often ignored intricacies that get lost in translation.
A: ebridge2 A lot of people work hands on with collaborators who collect data directly, however, no students that I know of in the lab at present seem to express interest in data acquisition and as such this hasn't really come up (most people are more interested in data processing/inference). Plenty of those data acquisition collaborators acquire new data at our direct recommendation, however, (ie, make decisions as to better data collection methods, or how they can improve their data collection to improve our inferential capacity) and the discussions are pretty hands-on even though our lab doesn't directly collect the data.
A: @BPedigo
I’m going to visit collaborators in england for several weeks for just this reason (and jovo has supported me going on international travel just to do this!). When I was working with an experimental lab at hopkins, I also went down to their lab to watch some experiments and ask questions about the way they were running things, and I think any experimental lab we collaborate with will be very open to this

Q: What is the process for adopting a project in the lab?

A: @BPedigo
I'm sure it varies a lot, Jovo didn't tell me what to work on though, and I had to try a few different things to figure out what I wanted to do. For me, the first week or two I was in the lab I asked Jovo to go through all of his grants with me. I found that helpful to get a big picture of what was going on, but it definitely didn't cover all of the options available to me. One project from the grants sounded interesting to me and I ended up meeting with that experimental lab, and I worked on that project off and on for a few months. I also met another experimentalist and I played with some analysis of his data that never really went anywhere, partially because I was working on the other project and taking classes. I also started a project called GraSPy while taking Jovo's class, which definitely was the most succesful out of those. Finally in April, I met with some ANOTHER group of experimentalists at a conference whom we had worked with previously, and I was really excited about their new data. I ended up deciding that was the project I wanted to focus on and I've been working on that ever since, and I'm using GraSPy to do a lot of the analysis, which is cool!
A: @hhelm
I came into the lab having completed a master's thesis under Carey Priebe (who works closely with the methods / statistical side of our lab). I planned on working on extensions of that research. This, however, was not obviously useful to our lab or for achieving our goals and so Jovo and I found a set projects for me that are both interesting and useful. In my experience, Jovo is able and willing to help you find meaningful (to you and the community) projects that are both in your area of expertise and well aligned with the lab's interests.
A: @j1c
Similarly, I spent first year during my Master's trying various projects. These ranged from contributing to algorithm development, computer vision projects (automatic synapse detection using conjugate array tomography data), and pipeline development. Currently, I have been working on GraSPY, and various macroscale connectomics projects.

Q: What do you think distinguishes the BME program at Hopkins from other programs?

A: @BPedigo
For me, one main thing was the huge freedom to take whatever classes you want. Other programs I was looking at would have made me take a few courses in things that had nothing to do with my research and interests, here, I have been able to take really good applied math courses that help my understanding of my work, we have acces to the medical school courses, it's really up to you. Other things are great access to collaborators, since there are amazing medical researchers and amazing engineers/mathematicians in the same place as adam said above. I also love having friends/colleagues in the PhD program that work on huge variety of topics, but are all really good at what they do, so you get a different perspective by talking to your peers about what you do and what they are working on.

Q: Does the lab have external collaborations at national labs or labs overseas?

A: @BPedigo
Yes. I work with a lab that just moved to England, I know of at least one in France and one in Germany. Probably way more than that. Less sure about national labs, I know Jovo has done work with some people at Argonne, probably others.

Q: What are your expectations from students just coming into the lab?

A: @jovo
Answer is here

Q : What is your favorite thing about being at JHU and in this group in particular?

A: @BPedigo
For JHU, I like having access to really smart collaborators/PIs/peers. Some of the best medical researchers, biomedical engineers, mathematicians, etc. It's really nice to be able to take classes and get to know those people. For the lab group, my answer is pretty similar. But to be more specific, I find our lab has a really good mix of talents. Some are really really good at math and stats, some at programming, some know more about the data applications we are working on, so there are always super knowledgable people that sit right around me that I can bounce ideas off of or ask questions. Everyone is very willing to help or offer an opinion. Also on top of that they're just all friendly so it makes for a fun work environment.

Q : What do PhD students do after they graduate?


Q : What is Jovo's mentoring style like?