Fig. 6
From: Machine learning for identification of silylated derivatives from mass spectra
A: Distribution of PubChem candidate set sizes for the test set spectra. Blue color represents the TMS derivatives that are in PubChem, and orange color represents the TMS derivatives that are not in PubChem; B: Distribution of TMS derivatives from test dataset; the y-axis represents the percentage of correct TMS derivatives not present in their candidate set (green color); percentage of correct TMS derivatives that are ranked top 1 (pink color); percentage of correct TMS derivatives ranked in the top 10 (yellow color); percentage of TMS derivatives that are ranked below the top 10 positions (orange color) and percentage of missing TMS derivatives, i.e., TMS derivatives that are not in the PubChem candidate set (blue color) among the test TMS derivatives with candidate set sizes in each size bin, while the x-axis represents the candidate set size bins.