IsoBayes

IsoBayes integrates mass spectrometry proteomics and transcriptomics data in a Bayesian probabilistic framework to enable inference at the single protein isoform level, including presence/absence detection, abundance estimation, and identification of transcript-protein abundance discrepancies.

Key Features:

• Bayesian probabilistic framework: Integrates mass spectrometry proteomics and transcriptomics data to enhance information for isoform-level inference.
• Two-layer latent variable approach: First samples whether a peptide has been correctly detected or filters peptides, then allocates the abundance of selected peptides across compatible proteins.
• Protein isoform inference: Infers presence or absence of each protein isoform via posterior probability.
• Abundance estimation: Estimates protein isoform abundance with credible intervals.
• Transcript-protein discrepancy detection: Targets isoforms where transcript and protein relative abundances significantly differ.

Scientific Applications:

• Single-isoform proteomics inference: Enables study of individual protein isoforms rather than abstracting results to gene-level or isoform groups.
• Peptide-to-protein inference: Recovers protein-level data from peptide-level data in bottom-up mass spectrometry proteomics.
• Multi-protease dataset analysis: Benchmarked and applicable to real datasets using multiple proteases.

Methodology:

IsoBayes employs a two-layer latent variable approach within a Bayesian probabilistic framework: first sampling whether a peptide has been correctly detected or filtering peptides, then allocating the abundance of selected peptides across the protein(s) they are compatible with, integrating mass spectrometry proteomics and transcriptomics data.