Improved intra-array and interarray normalization of peptide microarray phosphorylation for phosphorylome and kinome profiling by rational selection of relevant spots

Abstract

Massive parallel analysis using array technology has become the mainstay for analysis of genomes and transcriptomes. Analogously, the predominance of phosphorylation as a regulator of cellular metabolism has fostered the development of peptide arrays of kinase consensus substrates that allow the charting of cellular phosphorylation events (often called kinome profiling). However, whereas the bioinformatical framework for expression array analysis is well-developed, no advanced analysis tools are yet available for kinome profiling. Especially intra-array and interarray normalization of peptide array phosphorylation remain problematic, due to the absence of “housekeeping” kinases and the obvious fallacy of the assumption that different experimental conditions should exhibit equal amounts of kinase activity. Here we describe the development of analysis tools that reliably quantify phosphorylation of peptide arrays and that allow normalization of the signals obtained. We provide a method for intraslide gradient correction and spot quality control. We describe a novel interarray normalization procedure, named repetitive signal enhancement, RSE, which provides a mathematical approach to limit the false negative results occuring with the use of other normalization procedures. Using in silico and biological experiments we show that employing such protocols yields superior insight into cellular physiology as compared to classical analysis tools for kinome profiling.

Phosphoproteomics Analysis Identifies Novel Candidate Substrates of the Nonreceptor Tyrosine Kinase, Src-related Kinase Lacking C-terminal Regulatory Tyrosine and N-terminal Myristoylation Sites (SRMS)

“…Our analyses using peptide arrays revealed that the phosphorylation of several peptides could not be validated because these were rendered statistically insignificant. Importantly, it has been noted that because of technical reasons data normalization measures tend to account for the loss of information from nearly 50% to 70% of peptides printed on such arrays (58). Therefore, we normally use quasi-stringent t-testing (p value Ͻ 0.1) in our analyses to limit the bias associated with stringent statistical thresholds (81,82).…”

“…Therefore, we normally use quasi-stringent t-testing (p value Ͻ 0.1) in our analyses to limit the bias associated with stringent statistical thresholds (81,82). This method of applying a quasi-stringent t-testing threshold in peptide array analyses has also been described by Scholma et al (58).…”

“…The remaining two are control peptide, a negative control peptide corresponding to Beclin1 Y352 that was previously reported as a direct phosphorylation site for EGFR but not SRMS (57), and a peptide corresponding to the BRK Y447 residue because this was recently reported as a site that is directly phosphorylated by SRMS (11). Each peptide was printed in three triplicate sets (9 peptide replicates in total) on each array to allow for intra-array normalization (40,58) of the peptide phosphorylation intensities.…”

A Progressive Loss of phosphoSer138-Profilin Aligns with Symptomatic Course in the R6/2 Mouse Model of Huntington’s Disease: Possible Sex-Dependent Signaling

“…Further, given that a cellular phenotype is often the reflection of changes in the expression patterns of groups of signaling molecules with common biological functions, identifying a change in a group of these molecules is more biologically meaningful than a change in a single molecule. As importantly, it has been noted that 50–70% of the information from peptide arrays can be lost due to technical reasons during data normalization (Scholma et al 2016 ). The cut-off threshold used for InnateDB pathway analysis was more stringent ( P < 0.05 and FC > ± 1.5), with P -values being generated using the hypergeometric distribution test that confirms—prior to correction for multiple testing—whether a pathway is statistically more over-represented in the uploaded dataset than expected by chance.…”

An optimized framework for cancer prediction using immunosignature

“…These normalization methods are selected from the previous studies used on peptides, proteomics, and other microarray datasets. [ 12 15 16 17 ] These methods are described in the following subsection.…”