OBJECTIVES/SPECIFIC AIMS: To establish an effective team of researchers working towards developing and validating prognostic models employing use of image analyses and other numerical metadata to better understand pediatric undernutrition, and to learn how different approaches can be brought together collaboratively and efficiently. METHODS/STUDY POPULATION: Over the past 18 months we have established a transdisciplinary team spanning three countries and the Schools of Medicine, Engineering, Data Science and Global Health. We first identified two team leaders specifically a pediatric physician scientist (SS) and a data scientist/engineer (DB). The leaders worked together to recruit team members, with the understanding that different ideas are encouraged and will be used collaboratively to tackle the problem of pediatric undernutrition. The final data analytic and interpretative core team consisted of four data science students, two PhD students, an undergraduate biology major, a recent medical graduate, and a PhD research scientist. Additional collaborative members included faculty from Biomedical Engineering, the School of Medicine (Pediatrics and Pathology) along with international Global Health faculty from Pakistan and Zambia. We learned early on that it was important to understand what each of the member’s motivation for contributing to the project was along with aligning that motivation with the overall goals of the team. This made us help prioritize team member tasks and streamline ideas. We also incorporated a mechanism of weekly (monthly/bimonthly for global partners) meetings with informal oral presentations which consisted of each member’s current progress, thoughts and concerns, and next experimental goals. This method enabled team leaders to have a 3600 mechanism of feedback. Overall, we assessed the effectiveness of our team by two mechanisms: 1) ongoing team member feedback, including team leaders, and 2) progress of the research project. RESULTS/ANTICIPATED RESULTS: Our feedback has shown that on initial development of the team there was hesitance in communication due to the background diversity of our various member along with different cultural/social expectations. We used ice-breaking methods such as dedicated time for brief introductions, career directions, and life goals for each team member. We subsequently found that with the exception of one, all other team members noted our working environment professional and conducive to productivity. We also learnt from our method of ongoing constant feedback that at times, due to the complexity of different disciplines, some information was lost due to the difference in educational backgrounds. We have now employed new methods to relay information more effectively, with the use of not just sharing literature but also by explaining the content. The progress of our research project has varied over the past 4-6 months. There was a steep learning curve for almost every member, for example all the data science students had never studied anything related to medicine during their education, including minimal if none exposure to the ethics of medical research. Conversely, team members with medical/biology backgrounds had minimal prior exposure to computational modeling, computer engineering and the verbage of communicating mathematical algorithms. While this may have slowed our progress we learned that by asking questions and engaging every member it was easier to delegate tasks effectively. Once our team reached an overall understanding of each member’s goals there was a steady progress in the project, with new results and new methods of analysis being tested every week. DISCUSSION/SIGNIFICANCE OF IMPACT: We expect that our on-going collaboration will result in the development of new and novel modalities to understand and diagnose pediatric undernutrition, and can be used as a model to tackle several other problems. As with many team science projects, credit and authorship are challenges that we are outlining creative strategies for as suggested by International Committee of Medical Journal Editors (ICMJE) and other literature.
3165 Diseased and Healthy Gastrointestinal Tissue Data Mining requires an Engaged Transdisciplinary team