In a flipped classroom, students complete modules on their own that would traditionally be reserved for lectures outside of scheduled class time, therefore allowing for this time to be devoted to reinforcing student knowledge. Yet although the flipped classroom offers a compelling approach for fostering a constructivist, student-centric learning environment, research on the efficacy of flipped classes has been mixed. For that matter, is it even possible to successfully flip a classroom in an advanced, heavily specialized course? Over the past several years, I have implemented a flipped version of a “Fundamentals of Bioinformatics” course, along with the online materials that it leverages. In this talk, I will discuss some of the successes and pitfalls that I have encountered along the way.
11:00 AM-11:20 AM
Innovating life sciences education with computational modeling and simulations
Room: Columbus CD
Tomas Helikar, University of Nebraska-Lincoln, United States
Joseph Dauer, University of Nebraska - Lincoln, United States
Biological processes at almost every scale of biological organization are governed by complex, non-linear networks. Mathematical modeling and computer simulations have emerged as integral to life sciences research to understand these biological processes. Given the shift in life science research it is important for biology education to evolve in order to equip our students with skills to reason conceptually, mechanistically, and quantitatively, and to answer emerging life science questions. To address these challenges, we developed a new approach that enables students to learn through building, simulating, and investigating computational models of processes embedded in biological systems. This method is facilitated through an easy-to-use software, Cell Collective Learn (http://learn.cellcollective.org), that makes computational modeling accessible to any student and instructor (i.e., no prior computational modeling experience is necessary). We have developed and deployed lessons to cover a number of biological processes such as cell respiration, gene regulation, cell cycle, photosynthesis, glucose homeostasis, etc. This approach has been used at several levels, including large introductory courses, upper-level undergraduate, and graduate courses, as well as high school. The setting of its utility is also flexible; the modeling activities can be used in-class, assigned as homework, as well as deployed as extensive lab investigations.
The BLAST similarity searching programs are one of the most widely used, highly cited, and widely taught, software packages in Bioinformatics. If high school students, undergraduates, or graduate students, have any exposure to bioinformatics methods, they probably ran BLAST. But BLAST is not only a powerful tool for identifying related proteins and DNA sequences, it can also be used to illustrate fundamental concepts in biology, statistics, and computer science. We can expand the role of BLAST in Bioinformatics by shifting away from the traditional presentation of a BLAST "how-to" manual towards a more critical understanding of how BLAST works, and how its strengths and weaknesses can be used to illustrate evolutionary, statistical, and computational principles. BLAST is more than "Google for sequences." BLAST works both because related biological sequences share an evolutionary history, and because unrelated biological sequences do not share a history, and because alignments between unrelated sequences are very accurately modeled by the statistics of random strings of residues. Exploring the tension between significant and non-significant similarity, and closer and more distant homologs, can lead to a much deeper understanding of the relationships between biological phenomena with interrelated evolutionary histories and statistical approaches that assume independence.
11:40 AM-12:00 PM
Involving undergraduates in genomics research to narrow the education-research gap
Room: Columbus CD
Serghei Mangul, University of California, Los Angeles, United States
While the benefits of undergraduate research experiences (UREs) are recognized for
undergraduates, the advantages of UREs for graduate students, post-doctoral scholars, and
faculty are not clearly outlined. The analysis of genomic data is particularly well-suited for
successful involvement of undergraduates. We offer a framework for engaging
undergraduates in genomics research while simultaneously improving lab productivity. The
proposed strategy can be easily reproduced at other institutions, is pedagogically flexible,
and is scalable from smaller to larger laboratory sizes. We hope that genomics researchers
will involve undergraduates in more computational tasks that benefit both students and
senior laboratory members.
- Paper was published in Nature Biotechnology
12:00 PM-12:20 PM
A multidisciplinary wet-lab course for computational biology students
Room: Columbus CD
Joshua Kangas, Carnegie Mellon University, United States
Emily Furbee, University of Pittsburgh, United States
Karen Thickman, University of Washington, United States
Joseph Ayoob, University of Pittsburgh, United States
Graduate students in Computational Biology have strong quantitative backgrounds, but are often limited in their understanding of the theory, approach, and practice of biological experimentation. A strong grasp of the provenance of relevant biological data is essential for computational biologists to effectively critique and incorporate data into their research efforts. To give students this knowledge and insight, we have developed the Laboratory Methods for Computational Biologists (LMCB) course to provide a hands-on laboratory experience in four major areas: genomics, microscopy and bioimaging, high content screening, and x-ray crystallography. For each area, we have designed course modules covering general topics such as experimental design, limitations of common methods, cutting-edge and high throughput techniques, potential sources of error in data, and the preservation, analysis, and presentation of data. To provide the students with a more immersive research experience, we have designed the modules to cover one common research topic giving the students experience in using results from previous modules to inform the design of experiments for the next. The LMCB course provides foundational and experiential wet-lab training for the benefit of nascent computational scientists.
12:20 PM-12:40 PM
Proceedings Presentation: Proceedings: Training for translation between disciplines: a philosophy for life and data sciences curricula
Room: Columbus CD
K. Anton Feenstra, Vrije Universiteit Amsterdam, Netherlands
Motivation: Our society has become data-rich to the extent that research in many areas has become impossible without computational approaches. Educational programmes seem to be lagging behind this development. At the same time, there is a growing need not only for strong data-science skills, but foremost for the ability to both translate between tools and methods on the one hand, and application and problems on the other.
Results: Here we present our experiences with shaping and running a masters programme in Bioinformatics and Systems Biology in Amsterdam. From this, we have developed a comprehensive philosophy on how translation in training may be achieved in a dynamic and multidisciplinary research area, which is described here. We furthermore describe two requirements that enable translation, which we have found to be crucial: sufficient depth and focus on multidisciplinary topic areas, coupled with a balanced breadth from adjacent disciplines. Finally, we present concrete suggestions on how this may be implemented in practice, which may be relevant for the effectiveness of life science and data science curricula in general, and of particular interest to those who are in the process of setting up such curricula.
12:40 PM-2:00 PM
2:00 PM-2:40 PM
Keynote: Bioinformatics in the Undergraduate Classroom: Barriers to Integration
The Network for Integrating Bioinformatics into Life Sciences Education (NIBLSE) recently conducted a survey of more than 1200 undergraduate educators. We asked the respondents a number of open-ended questions seeking information about whether and to what extent they were able to incorporate bioinformatics into their classes. These responses were subjected to key word analysis and quantified.
Perhaps not surprisingly, the most frequently reported barrier was faculty training, reported by 24% of the respondents. We stratified the findings by a number of demographic variables. Results from this analysis showed that (1) faculty at Baccalaureate and Master’s institutions report this barrier at a higher rate than faculty at Associates or Doctoral institutions; (2) faculty who are themselves members of under-represented minorities reported this barrier at a higher rate (although the numbers were small); and (3) faculty who earned their terminal degrees most recently have more formal training than other cohorts but incorporate bioinformatics in their teaching less than other cohorts. These findings have implications for the prospects for integration of bioinformatics into the curriculum.
2:40 PM-3:00 PM
Implementing a competency-based training strategy for biomolecular researchers with high computational needs
Room: Columbus CD
Vera Matser, EMBL-EBI, United Kingdom
Rossen Apostolov, KTH Royal Institute of Technology, Sweden
Catherine Brooksbank, EMBL-EBI, United Kingdom
Adam Carter, The University of Edinburgh, United Kingdom
Lee Larcombe, nexaSTEM, United Kingdom
Arno Proeme, The University of Edinburgh, United Kingdom
BioExcel, an EU-funded Centre of Excellence for computational biomolecular research, has pioneered a training program based on a competency-based needs analysis. Competencies are sets of knowledge, skills and attitudes (KSAs); existing frameworks including ISCB’s were used to create the BioExcel profile. The competencies were mapped against existing training and new courses have been developed to fill specific gaps. The needs analysis highlighted that training resources related to High Performance Computing (HPC) are almost exclusively at intermediate and advanced level and require background knowledge of computing that many life scientists lack.
Using the competencies and KSAs, BioExcel has designed and delivered two training courses to address this gap. (1) Foundation skills for HPC in computational biomolecular research is a primer for people wanting to use HPC but who find the concept quite daunting and lack the needed computational skills. The course uses project-based learning and examples from day-to-day work (e.g. obtaining data, installing software), to empower life scientists to get the most out of using computers. (2) Hands-on Introduction to HPC for life scientists, focusses on basic HPC concepts, helping users judge how HPC can benefit their research, and equip them to efficiently make use of HPC (with PRACE).
3:00 PM-3:20 PM
Implementation and evaluation of different training modalities in resource limited settings
Room: Columbus CD
Nicola Mulder, University of Cape Town, South Africa
Shaun Aron, University of the Witwatersrand, South Africa
As the life sciences are continuing to generate ever larger and more complex datasets, the need for training in bioinformatics and data analysis has increased. In developing countries, in particular, bioinformatics training for computational biologists, life science researchers and health care professionals is in high demand. In resource limited settings such as in many African institutions, the high quality delivery of bioinformatics training is hampered by the small number of skilled trainers, slow Internet and poor access to reliable computing infrastructure. However, these limitations can be overcome by employing different modalities of training and delivery of training materials. H3ABioNet, a Pan African bioinformatics network, has had a strong focus on bioinformatics training for different audiences across Africa. Over the last 5 years, H3ABioNet has trained over 1,000 people through courses delivered over face to face workshops, hackathons, online courses and internships. Each training modality has its advantages and limitations, and their success depends on the required outcomes of the training and topic area. The impact of H3ABioNet training is monitored and evaluated through course evaluations and surveys. Here, we describe the different modalities used and evaluate and compare the impact of each on the training participants and their careers.
3:20 PM-3:40 PM
Mapping the interdisciplinary landscape to leverage community educational efforts
Room: Columbus CD
Shannon McWeeney, Oregon Health and Science University, United States
Ted Laderas, Oregon Health & Science University, United States
Justin Guinney, Sage Bionetworks, United States
William Hersh, Oregon Health & Science University, United States
Biomedical research is now inherently data intensive, with new technologies generating an increasingly large and diverse data landscape. We now have parallel needs to train people on how to best utilize these data, develop new approaches to solve key bottlenecks, visualize the results of complex analyses to make data “accessible”, and support reproducible analyses. A new national effort in data science education is being coordinated by the NCATS Center for Data to Health (CD2H) Education and Learning Innovation Working group (CD2H-ELI). Our mission is supporting the use and dissemination of cutting edge biomedical research informatics, computational biology and data science educational approaches, leveraging existing community efforts. Our first step in this work is an interdisciplinary landscape analysis to focus on assessment and harmonization of related computational and quantitative competencies. This is an inclusive, community-focused initiative which is focused on recognition, attribution and engagement with the existing efforts in this area by numerous groups e.g., ISCB, NLM, BD2K, CTSA, AMIA, ASA, etc.). We will discuss the current status, challenges faced in the harmonization process and the subsequent gap analysis under-addressed areas where training efforts and curricular development will need to be focused.
3:40 PM-4:00 PM
Invited Talk: Cognitive psychology in the bioinformatics learning enterprise
Room: Columbus CD
Allegra Via, Sapienza University, Rome, Italy
Teresa Attwood, The University of Manchester, United Kingdom
This presentation will describe and briefly discuss how specific ideas from the discipline of cognitive psychology can strengthen and guide the development, evaluation, and revision of teaching and learning opportunities in bioinformatics. We define the “bioinformatics learning enterprise” as a joint effort, by the instructor and the individual learner, to move the learner along a specific pre-articulated path of growth and development. We then discuss features of cognitive psychology, particularly those relevant for adult learners, that can be leveraged to promote engagement in “the learning enterprise”. A focus on cognitive psychological principles in the development, evaluation, and revision of teaching and learning opportunities in bioinformatics emphasizes the strengths this focus brings, and how it helps overcomes limitations that are inherent in competencies.
4:00 PM-4:40 PM
4:40 PM-5:00 PM
Invited Talk: Development of the NIBLSE Learning Resource Collection and Incubators
The Network for Integrating Bioinformatics into Life Sciences Education (NIBLSE) is an NSF-funded Research Coordination Network that aims to establish bioinformatics as an essential component of undergraduate life sciences education. As part of that effort, NIBLSE is working to make existing bioinformatics learning resources more accessible to non-specialists and to increase their use across undergraduate biology courses. To this end, NIBLSE has partnered with the Quantitative Undergraduate Biology Education and Synthesis (QUBES) project and CourseSource to develop and implement a novel model for supporting the refinement, publication, and dissemination of high quality bioinformatics teaching resources. NIBLSE Incubators are small, short-lived, online communities that work with an existing learning resource to (1) improve its usability across diverse life sciences classrooms, (2) introduce and teach important bioinformatics learning outcomes, and (3) move the learning resource toward publication and broader dissemination. This presentation will highlight the current status of the NIBLSE Learning Resource Collection and discuss the opportunities and challenges associated with the Incubator approach for refining learning resources.
NIBLSE and QUBES are supported by grants from the National Science Foundation (DBI 1539900 and DUE 1446269, respectively). Any opinions, findings, and conclusions or recommendations expressed in this material are those of the authors and do not necessarily reflect the views of the National Science Foundation.
5:00 PM-5:40 PM
Keynote: The Stanford Biomedical Informatics Curriculum: Early results from use in qualifying exams
The Stanford Biomedical Informatics (BMI) program has existed since 1982, granting PhD and MS degrees. The program continues to house students interested in both bioinformatics and clinical informatics (including imaging informatics), and the faculty has resisted specializing the program into only one of those areas, or splitting the program. The focus on methology development, validation and application makes allows these closely related application domains to co-exist within the program. In response to faculty concern that students were not sufficiently broad in their understanding of all areas of biomedical informatics (naturally tending to focus on the application domain and relevant methods, without broad synthesis of the field), we have recently reorganized our qualifying exam to be in two parts. In the first, students take a comprehensive written exam on the broad content of biomedical informatics, with a strong methodological bias. In the second, they write an NIH R01 style proposal outlining the specific aims of their research work. This all happens at the end of the second year. I will discuss the genesis of this new process, and review the broad categories for the written comprehensive exam which are: (1) modeling biomedical systems, (2) bioinformatics, (3) translational bioinformatics, (4) medical informatics, (5) imaging informatics, and (6) basic informatics research skills. The curriculum is still evolving and I will present a snapshot of our results using it in the first several qualifying examinations.
5:40 PM-6:00 PM
Education: Panel Discussion + Wrapup
Room: Columbus CD
Biography: Phillip Compeau
Phillip Compeau is the Assistant Department Head for Education in the Computational Biology Department, which is housed in Carnegie Mellon University’s School of Computer Science. He directs the Undergraduate Program in Computational Biology and serves as Assistant Director of the Master's in Computational Biology program (He holds a Ph.D. in mathematics He is the co-author (with Pavel Pevzner) of Bioinformatics Algorithms: An Active Learning Approach (http://bioinformaticsalgorithms.com/), which has been adopted by 65 institutions since 2015.
Phillip co-created (with Pavel Pevzner) the first massive open online course (MOOC) in bioinformatics, which has grown into the Bioinformatics Specialization on Coursera and received 200,000 enrollments since 2013. He also co-founded (with Nikolay Vyahhi) Rosalind, an online platform for learning bioinformatics that has hosted 50,000 active users.