The paper introduces SourceTracker, a Bayesian method for estimating the proportion of contaminants in microbial community samples from various environments. The method models contamination as a mixture of source communities in a sink community, with unknown mixing proportions. SourceTracker is applied to microbial surveys from neonatal intensive care units (NICUs), offices, and molecular biology laboratories, and a database of known contaminants is provided for future testing. The authors compare SourceTracker with other methods, such as naïve Bayes and random forests, demonstrating that SourceTracker performs better, especially in scenarios with low-coverage samples or when source environments lack a core set of taxa. The method is also effective in detecting low-level contamination and can be used in a wide range of microbial community studies, including forensic analysis and public health applications. The implementation of SourceTracker is available as an R package.The paper introduces SourceTracker, a Bayesian method for estimating the proportion of contaminants in microbial community samples from various environments. The method models contamination as a mixture of source communities in a sink community, with unknown mixing proportions. SourceTracker is applied to microbial surveys from neonatal intensive care units (NICUs), offices, and molecular biology laboratories, and a database of known contaminants is provided for future testing. The authors compare SourceTracker with other methods, such as naïve Bayes and random forests, demonstrating that SourceTracker performs better, especially in scenarios with low-coverage samples or when source environments lack a core set of taxa. The method is also effective in detecting low-level contamination and can be used in a wide range of microbial community studies, including forensic analysis and public health applications. The implementation of SourceTracker is available as an R package.