During one of MLA’s poster sessions, the title “Provision of a library bioinformatics service to support medical research” caught my eye. This poster explained how the Samuel J. Wood Library hired a research scientist and created services to support medical and research bioinformaticians at Weill Cornell Medical College (image included and used with permission). The poster offered specific details on the workshops, analytical support consultations, high-performance computers, and the scientific software hub. I was interested in learning more about the responsibilities, roles, and impact of this service, so I reached out to the authors, Terrie Wheeler, AMLS, and Peter Oxley, PhD, for the following interview:
Can you speak a bit about the level of collaboration between the research scientist and the librarians and/or library staff? Do any of the support responsibilities strictly belong to the research scientist or strictly to the librarians/staff?
Each collaboration is different in terms of involvement and roles. Sometimes, we [the librarians] are only running basic data quality checks and providing curation of the raw data files. Sometimes we [the librarians] are providing data visualization for results that have already been analyzed and published. It really depends on where the researcher is at when they first come to the library for assistance, and how much responsibility the researcher wishes to maintain for their project. Sometimes, we [the librarians] will even be asked for help regarding a specific analysis, but not be given access to their data! In which case, it becomes a teaching session about techniques and best practice, with the researcher/student taking full responsibility for their own work.
What was the most challenging aspect of establishing the library bioinformatics service?
There is an ongoing tug-of-war between being able to provide full, in-depth project support, and having the resources (mainly time) to support as many projects and people as possible. We are thus constantly recalibrating the types of service that add most value: quality control, data curation, and best practices for replicability and shareability are not often the priority or focus of the researcher. They are also the activities that a researcher is not likely to seek help for (and often are not seeking help for even when they come to the library), yet these things are of crucial importance in any experiment. So, we are trying to encourage everyone to avail themselves of our expertise in this, even when their first question is "how do I find which genes are important in my experiment?"
On a more logistical note, it continues to amaze people that bioinformatics support is available in the library, which says that our marketing still has a long way to go! But it is also connected to the general perception that libraries are just repositories for storing books.
How has offering this service helped increase usage of other library services?
This is a difficult question to answer precisely, as we have not surveyed users to see if the bioinformatics service has changed their perception or access of other library services. Anecdotally, we can assert that it has impacted some content access. Part of the bioinformatics service is to teach users about molecular databases and software that we support through subscription, and there are more questions from users about how to access this content following these workshops. The bioinformatics services themselves certainly feed into each other: running workshops not only educate the patrons about how to do bioinformatics, but also increases awareness of the consulting services we offer. Similarly, the consults offer an opportunity to determine where the need is for teaching, which informs our choice of workshop topics. We also are aware that some of our patrons consuming our bioinformatics service also consume our scientific software service.
What were the researchers’ options before the bioinformatics services were established?
Researchers at Weill Cornell Medicine have always been fortunate in terms of access to support. There is an Applied Bioinformatics Core, which helps labs undertake large bioinformatics projects, and has always hosted a drop-in consultation service once a week. A large number of postdoctoral bioinformaticians employed in labs have also served as an informal network of help and advice, though these human resources are quickly overwhelmed by the need to do their own research, let alone help everyone else! It has therefore been part of our role in the library, to learn about the expertise that already exists within the college, to enable people to connect to the best person for their particular question. It also means that we have a large knowledge base from which to draw when troubleshooting a particular problem.
In order to properly integrate the library bioinformatics service in our current environment, and complement, rather than compete with our Applied Bioinformatics Core, I [Terrie] included the Acting Director of the Applied Bioinformatics Core on the search committee for the bioinformaticist we hired [Peter]. We developed a concept for how our services would interface, and which each would champion. The library has acquired some advanced software that the Applied Bioinformatics Core recommended, we stay in touch with the other’s efforts, and refer patrons back and forth as appropriate.
Where can readers of this blog find more information on your topic?
We keep access to slides, notebooks, and bibliographies for many of our workshops online, which you can check out at github.com/oxpeter/library_bioinformatics_service
Special thanks to Terrie Wheeler, AMLS, and Peter Oxley, PhD for participating in this interview and for sharing their work!