Embedded Files
Overall aim : Chemical exposures and their health effects
Overall aim : Chemical exposures and their health effects
We focus on analyzing exposomics and metabolomics datasets in children's health, cancer, neurology and diabetes fields. We study how chemical exposures affect biological pathways in the human body and how those pathways can be linked with adverse phenotypes. We focus on chemical agents that have received high and medium priority by the national and international, governmental and research organizations.
Funding: We are supported by several NIH grants -
NIEHS Biomedical KnowledgeBases - U24ES035386 (MPIs Dinesh Barupal and Susan Teitelbaum) and R24ES036917 (MPIs Itai Kloog and Dinesh Barupal)
NIA - U01AG088562 (PI Rima Kaddurah-Douk, UT Health)
NCI - U24CA268153 (PI Susan Sumner, UNC)
NIEHS R01s - ES035478 (MPI Allison Kupsco and Dinesh Barupal), ES033688 (PI Dania Valvi), ES032831 (PI Doug Walker, Emory)
NIEHS P30 - P30ES023515 (PI Robert Wright)
NCATS CTSA - UL1TR004419 (PI Rosalind Wright)
main Technology GOALS
main Technology GOALS
1) To develop, integrate and implement software and databases for advancing exposomics/metabolomics data processing and analysis.
1) To develop, integrate and implement software and databases for advancing exposomics/metabolomics data processing and analysis.
This workflow can be applied on any scale of data generated using a GCMS or LCMS techniques ( low and high both resolutions). These tools and databases use chem-informatics, quantum chemistry, machine learning and bioinformatics algorithms and approaches.
2) To prioritize chemicals using NLP, biomedical text mining and database fusion.
2) To prioritize chemicals using NLP, biomedical text mining and database fusion.
We mine PubMed abstracts and PMC full-text data with text mining methods and combine several chemical databases to prioritize chemicals for exposome contexts.
2) To develop tools for mapping, visualizing and interpreting dys-regulated metabolic networks in different biological contexts.
2) To develop tools for mapping, visualizing and interpreting dys-regulated metabolic networks in different biological contexts.
Using biochemical and chemical relationships among chemicals, we map metabolic networks which can be used for identifying dys-regulated metabolic pathways and modules.
Computing Infrastructure
NVIDIA Node with 4 A100 GPUs
NVIDIA Node with 2 H100 GPUs
10 CPU nodes (overall 300 cores, 400TB storage)
Six High-Performance Workstations
Access to Mount Sinai Minerva HPC with 190 H100 GPUs and 30K CPU cores.
Commercial Software : TeraChem, NIST2026
Open-source/Free software : OpenSearch, DiffDock, Amber24, QUICK, QCxMS, ChemProp
Cloud Computing via NIH STRIDES Initiative
Page updated
Google Sites
Report abuse