DEVELOPMENT OF A ROBUST SCORE AND FALSE DISCOVERY RATE FOR METABOLITE IDENTIFICATION
EMSL Project ID
51431
Abstract
The most important part of any mass spectrometry-based metabolomics workflow is the ability to determine the matching quality between an unknown and its database match. In this project, we aim to use Bayesian statistical methods for estimating FDRs of small molecule identification and investigate the generalizability of these estimation methods. In addition, we aim to address current limitations in assignment of scores that will allow for accurate assignment of metabolites. The benefits to using Bayesian probability is that we can easily treat contributing parameters independently to generate individual probabilities, then combine those to generate a score that represents a probability of true identity.
Project Details
Start Date
2020-05-19
End Date
2020-09-30
Status
Closed
Released Data Link
Team
Principal Investigator