Clinical Translation of Protein Biomarkers Integrated with Bioinformatics

Abstract

Proteomics technology offers a conceptually attractive platform for disease biomarker discovery. In protein biomarker discovery phase, some maturely quantitative proteomics methods such as stable isotope labeling by amino acids in cell culture (SILAC), isobaric peptide tags for relative and absolute quantification (iTRAQ), and label-free can be chosen. A common endpoint for a biomarker discovery experiment is a list of putative marker proteins. A reasonable next step is to perform target quantitative measurements of these proteins in an expanded patient population to verify their statistical significance. In addition, developments in quantitative MS, such as multiple reaction monitoring (MRM), have greatly enhanced both the specificity and sensitivity of MS-based assays to the point that they can rival immunoassay for some analytes. It is powerful for a lot of candidates verified. But for clinical diagnosis or monitoring, in general, the target biomarker is less than five proteins, so an immunoassay such as ELISA, Western blot, and antibody microarray is more suitable for the detection.

Many of the experimental techniques are developed in proteomics, which allow lots of proteins can be detected or quantified simultaneously. At the same time, bioinformatics solutions being developed make us get the maximum information derived from proteome data sets. There are some protein identified pipelines: peptide mass fingerprinting (PMF), tandem MS, or MS/MS search with a theoretical digestion database sequences. The use of searching against a decoy database that comprises sequences known to be incorrect allows the false discovery rate (FDR) to be estimated; other methods such as spectral library approach and de novo peptide sequencing method can also be used for peptide identification. Quantitative proteomics can be separated into two major approaches: (1) the use of stable isotope labeling, use isobaric tags information in MS/MS or comparative 2D gel approach for quantitative, and (2) label-free techniques, use area under the curve (AUC) or signal intensity measurement based on precursor ion spectra or spectral counting, which is based on counting the number of peptides assigned to a protein in an MS/MS experiment. Databases and data processing methods help researchers discover and validate their interesting protein or biomarker much simpler.

With the bioinformatics and mass spectrometry technique developing, the clinical translation of protein biomarker is much easier than before. But effective biomarkers which can improve diagnosis, guide molecularly targeted therapy, and monitor activity and therapeutic responses across a wide spectrum of diseases are still difficult to get. The effective and quick translation of protein biomarkers needs the accelerated mass spectrometry-based biomarker researches and bioinformatics techniques exploring, as well as accelerated the development of such research applied to routine clinical tests, in order to achieve commercialization ultimately.