Normalized and Directional Interplay Scoring for the Interrogation of Proteoform Data

Abstract

Histone proteoforms, often presenting multiple co-occurring post-translational modifications (PTMs), are central to chromatin regulation and gene expression. A proteoform is a specific form of a protein that includes variations arising from genetic changes, alternative RNA splicing, proteolytic processing, and PTMs. Genomic context-dependent histone proteoforms define the histone code, influencing cellular phenotype by dictating interactions with DNA and chromatin-associated proteins. Understanding the dynamics of histone proteoforms is essential for elucidating chromatin-based regulatory mechanisms. Advances in middle-down and top-down proteomics methods enable accurate identification and quantitation of hundreds to thousands of proteoforms in a single run. However, the resulting data complexity presents significant challenges for analysis and visualization. Here, we introduce new computational methods to analyze the dynamics of histone PTMs and demonstrate their use in mouse organs during aging. We have developed and benchmarked two novel PTM crosstalk scores. The score that we term Normalized Interplay addresses limitations of the original crosstalk score Interplay providing a more complete and accurate measure of PTM crosstalk. The second score, delta I or Directional Interplay is an asymmetric measure quantifying the magnitude and directionality of crosstalk between PTMs. Applying our two-stage scoring approach to data from CrosstalkDB, a community resource that curates proteoform-level data, reveals the dynamics of histone H3 modifications during aging. The source code is available under an Apache license at https://github.com/k-p4/ptm_interplay_scoring.