Integrating Scientific Programming in Communities of Practice for Students in the Life Sciences

Research in life science domains is producing larger data sets that require the use of computational approaches to understand biological phenomena. Academic institutions, industry, and other sectors in the life sciences are creating jobs that involve computation, data science, and data visualization. Therefore, there is a need for life scientists to understand and be trained in computation for this new job market. Many life science students are not taught foundational concepts of computation as a part of their curriculum. Therefore, there exists a gap in understanding when beginning to learn computer science (CS) and relate it to data-centric questions in other fields. To improve learning experiences and help train these students, this work sought to understand existing challenges that life science students face in learning scientific programming and identify routes for improvement. To do so, we evaluated three distinct learning experiences--- a hands-on workshop, structured coursework, and long-term research experiences. Based on these student evaluations, we highlight the major challenges and benefits of different learning environments and provide suggestions to educators and institutions for integrating scientific programming education in life science coursework or research. Student-centered, group environments were the most successful at engaging students in computing concepts. Overall, this work provides strategies to enrich learning experiences and promote best practices in computation for life science students and engage these students in the development of in-demand skills.