Introduction

DiNovo is a software tool for automated, high-coverage and high-confidence de novo peptide sequencing from tandem mass spectrometry data based on multiple pairs of mirror proteases and deep learning technology. It is featured by:

  • Mirror-Spectra Recognition: Fast and accurate recognition of mirror spectral pairs based on a statistical scoring algorithm.
  • De Novo Sequencing: Peptide sequencing from mirror spectral pairs, including MirrorNovo algorithm (deep learning based, running on GPU), and pNovoM2 algorithm (updated version of pNovoM, graph theory based, running on CPU).
  • Quality Control: False discovery rate (FDR) estimation based on target-decoy strategies.
  • High Speed: Whole-process acceleration based on multi-level index system and optimized multiprocess parallel strategy.
  • Easy to Use: Friendly GUI design and optimized command-line interaction.

 

Software

DiNovo is written in Python3 and is available on GitHub:

   https://github.com/YanFuGroup/DiNovo

It can also be downloaded here:

   DiNovo binary release (Version 1.0.0 for Windows)   User guide

   DiNovo source code

 

Datasets

Please click the link below to download the example data to test DiNovo:

   Example data

 The training dataset of MirrorNovo are here below:

   Training Dataset of MirrorNovo

 

Publication

Zixuan Cao, Xueli Peng, Di Zhang, Piyu Zhou, Li Kang, Hao Chi, Ruitao Wu, Zhiyuan Cheng, Yao Zhang, Jiaxing Dai, Yanchang Li, Lijin Yao, Xinming Li, Yaoyu He, Jinghan Yang, Haipeng Wang, Ping Xu and Yan Fu. DiNovo enables high-coverage and high-confidence de novo peptide sequencing via mirror proteases and deep learning. Nature Communications, 17:2203, 2026.


Contact Us

Any problem with DiNovo or this website, please contact:

Prof. Yan Fu: yfu(at)amss(dot)ac(dot)cn