Researchers from Boston University have developed a teaching curriculum designed to help biomedical researchers use informatics in their biomedical research projects.
The course covers topics such as databases and data handling; data visualization; legal, regulatory, and ethical issues; and analysis of microarray data, and is designed to help translational researchers understand and communicate informatics issues relevant to their research, as well as apply informatics principles and methods to their projects, among other goals
In March, Robert Friedman, director of BU’s biomedical informatics division and one of the members of the development team member discussed the curriculum at the Summit on Translational Bioinformatics conference held in San Francisco, Calif.
Last month, BioInform spoke with two of members of the curriculum development team: Robert Friedman, director of BU’s biomedical informatics division and a professor of medicine, and Paola Sebastiani, a professor of biostatistics and bioinformatics at BU.
What follows is an edited version of the interview
I'd like to kick things off with some background on both of you.
RF: I am a physician. I've been in informatics since medical school which was really a long time ago. My major focus has been on translational clinical informatics. I've been in some administrative positions with respect to informatics here at BU and most specifically with the [BU Clinical and Translational Science Institute].
This particular project … came about a bit accidentally because education has not been my primary thing. It came about because of a solicitation from the [National Institutes of Health] that funds the [Clinical and Translational Science Award] program, and we proposed what we developed.
PS: I am a professor of biostatistics and bioinformatics at BU and I am also a member of the CTSI. I developed some training courses for the CTSI … I taught one or two day seminars focused on bioinformatics. That’s how I got involved in the project.
Have you published a paper providing details about the course that interested parties could take a look at?
RF: We will write a paper. The main reason it hasn’t been written is that generally speaking to have something published you have to have some 'results.' We are in the process of completing an evaluation of the program. I've had to encourage some of our study subjects who are all faculty at BU to endeavor to complete their evaluations of this even though at this point they're overdue about four or five months. We have complete results for around two-thirds of the people … and [those] that haven’t finished have done a lot of it. I expect that this will be written sometime in the spring or early summer.
I have presented it at a number of meetings — not only [TBI] but also at a CTSI-sponsored meeting. Over the past couple of years, I've probably presented it maybe four times.
You mentioned earlier that you submitted this project in response to a solicitation from the CTSA program. Can you tell me a little bit about the RFA and its goals?
RF: The CTSA program used to make announcements for applications from individual CTSA [institutes] for supplements to their grants. And they listed a whole bunch of topics that they were interested in. The one that I responded to … wasn’t specifically a call for informatics education, it wasn’t even a call for education, per se. It had to do with integrating different kinds of researchers so that they worked more effectively together. We morphed that into something which had the thesis of educating researchers across the spectrum about translational informatics principles and methods.
What's your take on existing translational informatics courses?
RF: Generally speaking, informatics education has been done by informaticians and has had the primary purpose of training more informaticians. Most of the courses that exist are more tilted to clinical informatics and population health informatics.
PS: I think there are typically workshops or tutorials that are offered before major conferences but they have a very specific target … they don’t reach the researchers. There is a lot of need for short … focused training.
RF: Research informatics did not really emerge as a sub-discipline of informatics until five or so years ago even though I date informatics to 1960. But research informatics that is focusing on how a researcher should consider using informatics in their research and having them come up to speed with their understanding of it so that they can be better researchers and they can also communicate with people who are dedicated experts. That’s a different thing.
Prior to what we were doing, all informatics education or training was focused on people who wanted to be informaticians, didn’t want to be researchers. Some people were researchers and also informaticians but they are a relatively small group. Up until five years ago, there wasn’t anything like this. Some of the CTSA [institutes] … have begun to develop courses in this area. Many of them are also directed at training future teachers, in other words training informaticians … to educate researchers in informatics. I think by now a number of CTSAs have some sort of program.
There is a lot of information that needs to be presented to the researchers. Can you talk a little bit about how you selected topics for the course and also how you came up with the structure you used?
RF: We did a formal needs assessment. We polled informaticians of all kinds who worked at the 60 CTSAs and in their CTSA work interacted with researchers [to help] them use informatics. [That] generally breaks itself down into about [three] different things: doing consultations; teaching actual seminars, courses, and so forth; [and] developing resources, computer systems, [and] software libraries for researchers at their institutions to use or to make them available … more generally. We had a very formal way of doing this which involved a number of online surveys and then we took the results, looked at them and said 'we agree with the decision of the group most of the time but a few cases we didn't mainly because some of the things they thought were important … were addressed within CTSAs or elsewhere. [For example] they were interested in teaching researchers how to access the biomedical literature using informatics query tools. [That's] very important but that’s something that all universities have and there are other venues for learning how to do it.
PS: For the actual modules on bioinformatics, which [focused on] the design and analysis of microarray experiments and next-generation sequencing technology, what I did was in part based on the requests that I receive regularly from investigators in the medical school. There was a need for training on the design and the analysis of these experiments. I have a full course over 15 weeks, so 45 hours [which] I condensed. I chose topics based on the priorities — what the largest demand [was] — but also it’s a collection of topics so that at the end of the training a person should feel confident to do some analysis by himself. There are other topics that show potential — [for example] what type of collaboration [does] an investigator [have] to establish to be able to extract full information from his experiment. [Also, there] is the topic of emerging technologies. So it was a combination of these needs that [was] the basis on which I chose the topics.
RF: One of the things we are doing now … is we have a diverse group of faculty who we have given access to the course, asked them to take it, and then asked them in a number of different ways to give us input about it in terms of general things and very specific things on how to make it better.
You said earlier that other CTSAs have started developing programs of their own. How's what you have done here different from what they are developing?
RF: I'd say most of the courses are traditional in the sense that they use face-to-face presentations, slides, and so forth. One of the major differences is that ours is totally online because we want to disseminate it to the other CTSAs. I think mostly what happens in course development is that people develop courses for seminar or classroom-type presentations and then adapt it for online presentation. We developed this course because of our understanding of the researchers and their time availability. One of the things about this is they can look at it anywhere [because] it's web-based. They could do it at their own pace, they could skip around…they don't have to actually do it in order.
PS: There is a level of professionalism in the modules that have been [developed]. So, its not like putting class notes on the internet. That’s not online teaching; it’s maybe the first step.
RF: Also, I think the future of it is it has to evolve over time. If we don’t constantly review it, update it with information from the students, update it with changes in the field that’s changing very rapidly … it won’t have any longevity.
How about in terms of content? Do you find that there is agreement between your course and courses that are being developed at other CTSAs?
PS: The modules that I put together, I would expect that the content is very similar. There are some established topics that you have to discuss if you want to work with microarray data. With next-generation sequencing technology it may be less so because this is an emerging technology so there are no established ways to do analysis yet.
RF: The most important thing is the selection of the topics. Then the other thing that’s really important is how much content is going to be in it because if you expect people to do it, then you are talking about time commitments. So you have to make some decisions there. The other thing [that] I think is the most demanding aspect of it is that our audiences are somewhat diverse.
We want to help the researcher understand enough that they can actually have a meaningful and productive consultation with an expert about their work. They need to know enough about the principles and, ideally, they should be able to critique their own research and already have ideas where they can use the methods in their research. We are trying to achieve that. [Also, they need] to be able to evolve with the field over time. The third thing … is the researcher has to be the [sternest] critique of his research design and methods from the perspective of the discipline.
How are you planning to disseminate this course more generally?
RF: We built that into the original grant. What we originally said was we are going to disseminate it through the CTSA program. We could disseminate in a read-only version [and] we could disseminate it [in a form] they could modify, which in some ways is actually necessary. Then the question [is]: what part of the university would use it; is it the CTSA? Is it the informatics department? Maybe it's some subset.
There are options. I've promised people who were involved in this that we would disseminate it so they can use it. More than that I can't tell you. The world is moving so rapidly that it's not an easy question to answer.