代谢组学
化学
前列腺癌
生物标志物发现
质谱法
数据集
色谱法
串联质谱法
人工智能
模式识别(心理学)
计算机科学
癌症
蛋白质组学
内科学
基因
医学
生物化学
作者
Francisco de Assis de Carvalho Pinto,Iqbal Mahmud,Vanessa Rubio,Ademar Domingos Viagem Máquina,Anízia Fausta Furtado Durans,Waldomiro Borges Neto,Timothy J. Garrett
标识
DOI:10.1021/acs.analchem.1c04004
摘要
Sensitive, rapid, and meaningful diagnostic tools for prostate cancer (PC) screening are urgently needed. Paper spray ionization mass spectrometry (PSI-MS) is an emerging rapid technology for detecting biomarker and disease diagnoses. Due to lack of chromatography and difficulties in employing tandem MS, PSI-MS-based untargeted metabolomics often suffers from increased ion suppression and subsequent feature detection, affecting chemometric methods for disease classification. This study first evaluated the data-driven soft independent modeling of class analogy (DD-SIMCA) model to analyze PSI-MS-based global metabolomics of a urine data matrix to classify PC. The efficiency of DD-SIMCA was analyzed based on the sensitivity and specificity parameters that showed 100% correct classification of the training set, based on only PC and test set samples, based on normal and PC. This analytical methodology is easy to interpret and efficient and does not require any prior information from the healthy individual. This new application of DD-SIMCA in PSI-MS-based metabolomics for PC disease classification could also be extended to other diseases and opens a rapid strategy to discriminate against health problems.
科研通智能强力驱动
Strongly Powered by AbleSci AI