- October 18 - 20, 2023
- - ET
- - ET
- - ET
Event LocationWhite Oak Campus, Great Room A
10903 New Hampshire Avenue
Silver Spring, MD 20993
- FDA Modeling and Simulation Working Group
- Chemical Informatics and Modeling Interest Group
- Women of CDER, Artificial Intelligence Interest Group
- EPA, Center for Computational Toxicology and Exposure
- NIH, Frederick National Laboratory for Cancer Research (FNLCR), Cancer Data Science Initiatives
- NIH, National Center for Biotechnology Information, PubChem
- NIST, Mass Spectrometry Data Center
Location: FDA White Oak campus, Great Room, and Zoom
FDA’s Chemical Informatics and Modeling Interest Group, together with Women of CDER, Artificial Intelligence Interest Group, are hosting a workshop for Government-funded organizations on October 18-20, 2023, from 9:00 a.m.–4:30 p.m. ET.
The workshop Cheminformatics Resources of U.S. Governmental Organizations will bring together cheminformaticians working in different branches of the U.S. Government to share experience and challenges in creating and maintaining computational resources pertaining to structures and properties of molecules and materials.
About the Workshop
Session: The Path Forward for Machine Learning with Chemical Data and Molecular Representations
Session chairs: Dr. Eric A Stahlberg (FNLCR/Cancer Data Science Initiatives), Dr. Samir Lababidi (FDA/OC/Office of Data, Analytics, and Research)
The session will bring together the disciplines of data science and cheminformatics to share current capabilities and establish future directions for collaborative efforts in the development of machine learning models involving chemical information. Topics covered include exploring the state and opportunities for common representations of chemical information, potential conventions for molecular representations, and requirements for greater model portability and interoperability. The session will also involve presentations on workflows and frameworks that can be used to help compare models; explore efforts where chemical data are combined with other data to develop computational models; and involve discussion on current molecular representations as well as possible improvements needed in future representations to support machine learning and ultra-large chemical databases. The session concludes with an interactive panel discussion on topics of interest to attendees.
Session: Application of Cheminformatics to Support Analytical Chemistry
Session chairs: Dr. Antony J. Williams (U.S.-EPA/Center for Computational Toxicology and Exposure), Dr. Tytus Mak (NIST/Mass Spectrometry Data Center)
This session will explore the various applications of cheminformatics in analytical chemistry, including data processing, analysis, and visualization and challenges in integrating cheminformatics into analytical chemistry workflows. Topics of discussion will include the use of cheminformatics to develop machine learning models for chemical analysis (e.g., retention time prediction, method amenability, mass spectrometry fragmentation), the interpretation of spectral data, the design and optimization of chromatographic separations, and the integration of cheminformatics tools into laboratory information management systems supporting analytical chemistry. Attendees will gain a deeper understanding of the potential benefits and challenges of using cheminformatics to support analytical chemistry, and will leave with practical insights for incorporating these tools into their own research and analysis workflows.
Session: FAIR-ifying and Sharing Chemical-Related Data
Session chairs: Dr. Evan Bolton (NIH/NCBI/PubChem), Dr. Huixiao Hong (FDA/NCTR/Division of Bioinformatics and Biostatistics)
By now, we have all heard of FAIR. Depending on your perspective, you may think of it as “Findable, Accessible, Interoperable, and Reusable,” or, to some, “Fully AI Ready.” FAIR-ifying (i.e., the process of making data FAIR) often means adding sufficient metadata and structure to data such that it is (more) machine-readable. Many of us produce chemical-related data. Many of us use chemical-related data. What are the best practices for FAIR-ifying chemical-related data? What are your pain points in (re)using chemical-related data? As federal employees, how can we improve the chemical-related data ecosystem? How can we work together to improve chemical-related data-sharing as a whole? This session will explore these questions and more.
|Meeting Agenda||pdf (152.22 KB)|