Aerith: Visualization and Annotation of Isotopic Enrichment Patterns of Peptides and Metabolites with Stable Isotope Labeling from Proteomics and Metabolomics

TL;DR

Aerith is a software tool that helps scientists visualize and analyze isotopic labeling patterns in peptides and metabolites using mass spectrometry data.

Contribution

Aerith introduces efficient algorithms and visualization tools for accurate simulation and annotation of isotopic patterns in SIP studies.

Findings

Aerith uses convolution and Monte Carlo simulations to model isotopic distributions efficiently.

The tool enables precise annotation and manual validation of isotopically labeled peptides and metabolites.

Case studies demonstrate its utility in resolving isotopic structures for compounds like glucose and penicillin.

Abstract

Stable isotope probing (SIP) traces the metabolism of
biological
cells using isotopically heavy substrates (e.g., 13C, 15N, or 2H). Confident identification of the metabolic
products of isotopic labeling remains a challenge due to the difficulties
in simulating, visualizing and annotating the isotopic patterns of
partially labeled peptides and metabolites found in mass spectrometry
(MS) data. Here, we present Aerith, an R package designed to visualize
data of simulated and observed isotopic envelopes of peptides and
metabolites with user-defined formula and atom % enrichment levels.
Aerith models the isotopic distributions of the fragment ion series
of a peptide by sequentially convoluting isotopic envelopes of monomeric
units using a convolution algorithm. Aerith simulates fine isotopic
structures of a compound using Monte Carlo simulation via the multinomial
distribution, and calculates the isotopic envelopes of metabolites
with known chemical formulas using an FFT-based algorithm. These algorithms
provide accurate simulation of the isotopic envelopes of SIP-labeled
peptides and metabolites with high computational efficiency. Aerith
evaluates peptide-spectrum matches through multiple robust and commonly
used scoring functions to compare experimental and theoretical spectra.
These algorithms were implemented in C++ and accessed in R via Rcpp
to ensure real-time interactivity and significantly improve computational
efficiency compared to native R code. We present case studies to demonstrate
Aerith’s utility in resolving isotopic fine structures and
envelopes for glucose, penicillin, and microbial peptides containing
natural and enriched isotopes. By providing visualization of isotopically
labeled peptides and metabolites, Aerith enables precise annotation
of their mass spectra and manual validation of their identifications
in proteomic and metabolomic SIP studies.

Figures