Fig. 1: Algorithmic workflow and modules of DiNovo. | Nature Communications

Fig. 1: Algorithmic workflow and modules of DiNovo.

From: DiNovo enables high-coverage and high-confidence de novo peptide sequencing via mirror proteases and deep learning

Fig. 1: Algorithmic workflow and modules of DiNovo.The alternative text for this image may have been generated using AI.

a The workflow of DiNovo software. Proteome samples are first digested using multiple mirror proteases, followed by LC-MS/MS analysis to obtain tandem mass spectra. Then, mirror spectral pairs are recognized by the MirrorFinder algorithm, and peptides are sequenced by the MirrorNovo or pNovoM2 algorithm. b MirrorFinder, the mirror-spectra recognition module of DiNovo, which leverages the distribution of fragment ion mass differences (delta masses) between two spectra to recognize mirror spectra and employs a target-decoy strategy for recognition error control. c MirrorNovo, the prime de novo sequencing module of DiNovo, which makes use of a designed neural network architecture to capture the peak features of mirror spectra and predict the probabilities of amino acids.

Back to article page