PSB Workshops

Large Language Models (LLMs) are a type of artificial intelligence that has been revolutionizing various fields, including biomedicine. They have the capability to process and analyze large amounts of data, understand natural language, and generate new content, making them highly useful in many applications. In this workshop, we aim to introduce the attendees to an in-depth understanding of the role of LLMs in biomedicine (e.g., in research, clinical informatics, education, and ethics), and how they are being used to drive innovation and improve outcomes in the field.

Contact: Zhiyong Lu
Email: zhiyonglu at nih dot gov

In biomedical research and clinical medicine, many of the ethical frameworks and processes focus on benefits and harms at the individual level. However, in biomedicine, there is increasing recognition of a need to implement frameworks and processes that address the social impacts of technologies, such as genomics and AI technologies, and their social benefit for underrepresented populations and communities. For example, studies demonstrating the potential for bias in AI shed light on the need to develop processes to more effectively identify and address downstream impacts of medical AI, as well as engage communities who are stakeholders in the research. Privacy is often envisioned as an individual right, but the collection and use of data also have repercussions at the level of groups and communities. For that reason, there have been recent efforts to arrive at models for data stewardship and data sovereignty. This workshop will provide a forum for discussion of practical approaches to enhancing fairness, social responsibility and inclusion of diverse viewpoints in biomedicine. Interdisciplinary research on ethics and how fairness, social responsibilit,y and community engagement can be operationalized in biomedical research will provide a foundation for robust discussion on these issues.

Contact: Nicole Martinez-Martin
Email: nicolemz at stanford.edu

The primary efforts of disease and epidemiological research can be divided into two areas: identifying the causal mechanisms and utilizing important variables for risk prediction. The latter is generally perceived as a more obtainable goal due to the vast number of readily available tools and the faster pace of obtaining results. However, the lower barrier of entry in risk prediction means that it is easy to make predictions, yet it is incredibility more difficult to make sound predictions. As an ever-growing amount of data is being generated, developing risk prediction models and turning them into clinically actionable findings is crucial as the next step. However, there are still sizable gaps before risk prediction models can be implemented clinically. While clinicians are eager to embrace new ways to improve patients’ care, they are overwhelmed by a plethora of prediction methods. Thus, the next generation of prediction models will need to shift from making simple predictions towards interpretable, equitable, explainable and ultimately, casual predictions. The purpose of this workshop is to introduce and discuss the current and future of risk prediction in the context of disease and epidemiological research. We will discuss the pressing topics ranging from data sources to model implementation.

Contact: Ruowang Li
Email: ruowang.li at cshs.org

Immune modulation is considered a hallmark of cancer initiation and progression and immune cell density has been consistently associated with outcomes of cancer patients. Multiplex immunofluorescence (mIF) microscopy combined with automated image analysis is a novel and increasingly used technique that allows for the assessment and visualization of the tumor immune microenvironment (TIME). Recently, application of this new technology to tissue microarrays (TMAs) or whole tissue sections from large cancer studies has been used to characterize different cell populations in the TIME with enhanced reproducibility and accuracy. Generally, mIF data has been used to examine the presence and abundance of immune cells in the tumor; however, this aggregate measure assumes uniform patterns of immune cells throughout the tumor and overlooks spatial heterogeneity. Recently, the spatial contexture of the TIME has been explored with a variety of methods. In this session, speakers will present some of the state-of-the-art statistical methods for assessing the TIME from mIF data.

Contact: Brooke Fridley
Email: brooke.fridley at moffitt dot org

Cells consists of large components, such as organelles, that recursively factor into smaller systems, such as condensates and protein complexes, forming an intricate time-dependent multi-scale structure of the cell. Remarkable molecular profiling technologies, such high throughput affinity co-purification, that emerged in the past decade are beginning to be complemented by new technologies, such as high throughout protein fluorescent imaging and cryo-ET, that systematically interrogate the subcellular structure of cells. In this workshop, we will discuss progress, challenges, and collaboration to marshal various computational approaches towards assembling an integrated structural map of the human cell.

Contact: Emma Lundberg
Email: emmalu at stanford.edu