Graph Representations and Algorithms in Biomedicine
PSB 2023
Session Overview
Connectivity is a fundamental property of biological systems: on the cellular level, proteins interact with each other to form protein-protein interaction networks; on the organism level, neurons are arranged in a network; and on a community level, species can have complex relationships with one another that drive the development of an ecosystem. Graphs, mathematical representations modeling entities as vertices and their relationships as edges, have proved useful for understanding biological systems that naturally have such a network structure. Graph representations and algorithms (often in combination with machine learning techniques) can be used to organize massive amounts of related (and sometimes heterogenous or unstructured) data, and to ultimately to identify patterns that represent novel biological insights. Additionally, recent advances in high order networks, hypergraphs, and computational topology promise to bring a higher level of complexity to network models. This PSB session ``Graph Representations and Algorithms in Biomedicine,'' will encompass modern developments in graph theory, computational topology, and graph machine learning applied to various fields of biomedicine.
Session Topics
We invite submissions that aim to advance biomedicine by constructing, comparing, and analyzing graphs and hypergraphs. Broadly, this may include:
Algorithms for assembling graphs from large datasets and available metadata
Graph algorithms for processing heterogeneous and unstructured datasets
Novel machine learning techniques for node, link, and graph prediction tasks
Generative models for constructing graphs
Applications of graphs to drug design, precision health, clinical practice, and biomedical ontologies
Advancing fairness and interpretability of graph-based algorithms in biomedicine
Specific topics of interest include:
Longitudinal graph-based analysis for temporal data
Hybrid graph and genome-based approaches for population genetics
Biomedical knowledge graph construction from large and heterogeneous corpora of data (e.g. multi-omics, EHR, biomedical text, etc.)
Graph representation learning algorithms for generating node embeddings
Interpretability of graph neural networks
Measurements and methods for ensuring fairness of graph algorithms
Visualization techniques for graph algorithms
Session Organizers
Brianna Chrisman
PhD Student
Bioengineering
Stanford University
Cliff Joslyn
Chief Knowledge Scientist and
Mathematics of Data Science Lead
Pacific Northwest
National Laboratory
Maya Varma
PhD Student
Computer Science
Stanford University
Sepideh Maleki
PhD Student
Computer Science
University of Texas, Austin
Maria Brbic
Postdoctoral Researcher
Computer Science
Stanford University
Marinka Zitnik
Assistant Professor of Biomedical Informatics
Harvard University
Submission Information
All submitted papers are fully reviewed and accepted on a competitive basis.
Important dates
August 1, 2022: Paper submissions due
September 9, 2022: Notification of paper acceptance
October 3, 2022: Camera-ready paper deadline
December 5, 2022: Abstract deadline for posters
January 3-7, 2023: Conference dates
Submission Information
Please follow submission guidelines at psb.stanford.edu.