High-throughput proteomics and AI for cancer biomarker discovery

Biomarkers are assayed to assess biological and pathological status. Recent advances in high-throughput proteomic technology provide opportunities for developing next generation biomarkers for clinical practice aided by artificial intelligence (AI) based techniques. We summarize the advances and limitations of cancer biomarkers based on genomic and transcriptomic analysis, as well as classical antibody-based methodologies. Then we review recent progresses in mass spectrometry (MS)-based proteomics in terms of sample preparation, peptide fractionation by liquid chromatography (LC) and mass spectrometric data acquisition. We highlight applications of AI techniques in high-throughput clinical studies as compared with clinical decisions based on singular features. This review sets out our approach for discovering clinical biomarkers in studies using proteomic big data technology conjoined with computational and statistical methods. (c) 2021 Elsevier B.V. All rights reserved.

Automated summary

This paper summarizes the advances and limitations of biomarkers in cancer research, as well as the important applications of mass spectrometry technology in proteomics. The article emphasizes the role of artificial intelligence technology in clinical studies, as well as the importance of combining computational and statistical methods in biomarker research.