Abstract
AbstractSelected-ion flow-tube mass spectrometry, SIFT-MS, technology seems nowadays very promising to be utilized for the discovery and profiling of biomarkers such as volatile compounds, trace gases, and proteins from biological and clinical samples. A high performance biomarker detection method for identifying biomarkers across experimental groups is proposed for the analysis of SIFT-MS mass spectrometry data. Analysis of mass spectrometry data is often complex due to experimental design. Although several methods have been proposed for the identification of biomarkers from mass spectrometry data, there has been only a handful of methods for SIFT-MS data. Our detection method entails a three-step process that facilitates a comprehensive screening of the mass spectrometry data. First, raw mass spectrometry data are pre-processed to capture true biological signal. Second, the pre-processed data are screened via a random-forest-based screening tool. Finally, a visualization tool is complementing the findings from the previous step. In this paper, we present two applications of our method; a control-asthma case study and an H1N1 Flumist time-course case study.KeywordsBiomarkerRandom forestMass spectrometrySIFT-MS
Sign-in/Register to access full text options 
Free) 
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
