Abstract
Various studies have shown associations between molecular features and phenotypes of biological samples. These studies, however, focus on a single phenotype per study and are not applicable to repository scale metabolomics data. Here we report MetSummarizer, a method for predicting (i) the biological phenotypes of environmental and host-oriented samples, and (ii) the raw ingredient composition of complex mixtures. We show that the aggregation of various metabolomic datasets can improve the accuracy of predictions. Since these datasets have been collected using different standards at various laboratories, in order to get unbiased results it is crucial to detect and discard standard-specific features during the classification step. We further report high accuracy in prediction of the raw ingredient composition of complex foods from the Global Foodomics Project.
Highlights
Ingredients whose combination is most similar to the query
MetClassifier predicts the phenotype of samples in the following steps (Fig. 1A): (i) reference mass spectra of environmental/host-oriented samples are collected, (ii) mass spectrometry feature and phenotype metadata matrices are formed, (iii) logistic regression classifier is trained for predicting phenotype from mass spectrometry features, (iv) mass spectrometry feature vector is formed for query sample, and (v) the classifier predicts phenotypes for query sample
Metabolomic data collected from various body-sites has the potential for revealing the molecular mechanism of disease, providing the path toward diagnosis
Summary
Ingredients whose combination is most similar to the query. Benchmarking our method on data from the Global FoodOmics Project[3,4,5,6,7,8,9,10] shows a remarkable consistency between the ingredients reported by MetSummarizer and the known ingredients.
Sign-in/Register to view highlights & summary Sign-in/Register to access full text options 
Talk to us
Join us for a 30 min session where you can share your feedback and ask us any queries you have
Schedule a call
