NovoHMM

NovoHMM employs a generative hidden Markov model and Bayesian inference to perform de novo peptide sequencing from mass spectrometry data by estimating posterior probabilities for amino acids.

Key Features:

• Generative Hidden Markov Model: Employs a generative HMM tailored for de novo peptide sequencing from mass spectrometry data using a probabilistic framework.
• Graphical and Factorial Extensions: Extends the base HMM into a graphical model and a factorial HMM to model relationships between amino acids and spectral observations.
• Posterior Probability Estimation: Estimates posterior probabilities for amino acids within peptide sequences under a Bayesian framework rather than assigning independent symbol scores.

Scientific Applications:

• De novo peptide sequencing: Identification of peptide sequences directly from mass spectrometry data without relying on sequence databases.
• Proteome characterization: Interpretation of complex mass spectrometry data to inform protein composition and amino acid-level inference.
• Large-scale proteomic studies: Application to high-throughput proteomic datasets where probabilistic inference of peptide sequences is required.

Methodology:

Implements a Bayesian framework integrating a generative HMM with graphical and factorial HMM extensions to estimate posterior probabilities of amino acids from mass spectrometry data.