Selected article for: "cross validation and error rate"

Author: Marcus Ludwig; Louis-Félix Nothias; Kai Dührkop; Irina Koester; Markus Fleischauer; Martin A. Hoffmann; Daniel Petras; Fernando Vargas; Mustafa Morsy; Lihini Aluwihare; Pieter C. Dorrestein; Sebastian Böcker
Title: ZODIAC: database-independent molecular formula annotation using Gibbs sampling reveals unknown small molecules
  • Document date: 2019_11_16
  • ID: 03uonbrv_7
    Snippet: Some compounds in an LC-MS/MS run can result in high-scoring hits when searching in a MS/MS spectral library. ZODIAC's stochastic model allows us to integrate these hits as anchors, assuming that we can trust assigned molecular formulas to a high degree. We performed a 10-fold cross-validation to assess the improvement using anchors. We ensured structure-disjoint evaluation on the library hits, as multiple compounds in the dataset may correspond .....
    Document: Some compounds in an LC-MS/MS run can result in high-scoring hits when searching in a MS/MS spectral library. ZODIAC's stochastic model allows us to integrate these hits as anchors, assuming that we can trust assigned molecular formulas to a high degree. We performed a 10-fold cross-validation to assess the improvement using anchors. We ensured structure-disjoint evaluation on the library hits, as multiple compounds in the dataset may correspond to the same structure; see ref. 8 on the importance of structure-disjoint evaluation. ZODIAC with anchors improves the error rate on the tomato dataset to 1.48 % and on mice stool to 18.60 %, but does not improve results for the other datasets.

    Search related documents:
    Co phrase search for related documents
    • Try single phrases listed below for: 1